{ "cells": [ { "cell_type": "code", "execution_count": 1, "id": "55e1d103", "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "PyTorch version: 2.2.2\n", "CUDA available: False\n", "0.0.5\n" ] } ], "source": [ "import time\n", "from pathlib import Path\n", "\n", "import numpy as np\n", "import pandas as pd\n", "import matplotlib.pyplot as plt\n", "import seaborn as sns\n", "from scipy import stats\n", "\n", "from sklearn.model_selection import KFold\n", "from sklearn.pipeline import make_pipeline\n", "from sklearn.impute import SimpleImputer\n", "from sklearn.preprocessing import StandardScaler\n", "from sklearn.ensemble import RandomForestRegressor\n", "\n", "import torch\n", "print(f\"PyTorch version: {torch.__version__}\")\n", "print(f\"CUDA available: {torch.cuda.is_available()}\")\n", "\n", "import fdfi\n", "print(fdfi.__version__)\n", "\n", "from fdfi.explainers import FlowExplainer, Crossfitting" ] }, { "cell_type": "markdown", "id": "9377c8e2", "metadata": {}, "source": [ "## Load Data and Feature Groups\n" ] }, { "cell_type": "code", "execution_count": 2, "id": "60bcf1d9", "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "Dataset shape: (611, 832)\n", "Target variable: sens50\n", "Number of groups: 14\n", "Data loaded from: data/sens50_processed_dataset.csv\n" ] } ], "source": [ "DATA_DIR = Path(\"data\")\n", "\n", "DATASETS = {\n", " \"sens50\": {\n", " \"file\": DATA_DIR / \"sens50_processed_dataset.csv\",\n", " \"target\": \"sens50\",\n", " },\n", " \"sens80\": {\n", " \"file\": DATA_DIR / \"sens80_processed_dataset.csv\",\n", " \"target\": \"sens80\",\n", " },\n", " \"ic50\": {\n", " \"file\": DATA_DIR / \"ic50.censored_processed_dataset.csv\",\n", " \"target\": \"ic50.censored\",\n", " },\n", "}\n", "\n", "outcome = \"sens50\"\n", "\n", "cfg = DATASETS[outcome]\n", "filename = str(cfg[\"file\"])\n", "\n", "data = pd.read_csv(cfg[\"file\"])\n", "y = data[cfg[\"target\"]].astype(int)\n", "X = data.drop(columns=[cfg[\"target\"]])\n", "\n", "df_groups = pd.read_csv(DATA_DIR / \"feature_group.csv\", index_col=0).set_index(\"feature\")\n", "df_groups = pd.get_dummies(df_groups[\"group\"]).groupby(df_groups.index).max().reindex(X.columns)\n", "\n", "print(f\"Dataset shape: {X.shape}\")\n", "print(f\"Target variable: {outcome}\")\n", "print(f\"Number of groups: {len(df_groups.columns)}\")\n", "print(f\"Data loaded from: {filename}\")\n" ] }, { "cell_type": "code", "execution_count": 3, "id": "2df72507", "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "Data preparation...\n", "Data preparation finished\n", "Feature matrix shape: (611, 832)\n", "y statistics: mean=-0.000, std=1.000\n", "Missing values - X: 0, y: 0\n", "Data preparation finished\n", "Feature matrix shape: (611, 832)\n", "y statistics: mean=-0.000, std=1.000\n", "Missing values - X: 0, y: 0\n" ] } ], "source": [ "print(\"Data preparation...\")\n", "\n", "preprocess = make_pipeline(\n", " SimpleImputer(strategy=\"most_frequent\"),\n", " StandardScaler(),\n", ")\n", "\n", "X_values = preprocess.fit_transform(X)\n", "y_values = y.values.astype(float)\n", "\n", "# Standardize y to N(0,1) — matches the EOT notebook setup\n", "y_scaler = StandardScaler()\n", "y_values = y_scaler.fit_transform(y_values.reshape(-1, 1)).ravel()\n", "\n", "print(\"Data preparation finished\")\n", "print(f\"Feature matrix shape: {X_values.shape}\")\n", "print(f\"y statistics: mean={y_values.mean():.3f}, std={y_values.std():.3f}\")\n", "print(f\"Missing values - X: {np.isnan(X_values).sum()}, y: {np.isnan(y_values).sum()}\")\n" ] }, { "cell_type": "code", "execution_count": 4, "id": "a6e75e1c", "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "RandomForestRegressor fitted on full data\n", "Train R²: 0.705\n" ] } ], "source": [ "# Fit RandomForestRegressor on FULL data (matching DFI paper: refit_mu=False).\n", "rf = RandomForestRegressor(\n", " n_estimators=500,\n", " max_depth=None,\n", " min_samples_leaf=5,\n", " random_state=0,\n", " n_jobs=-1,\n", ")\n", "rf.fit(X_values, y_values)\n", "\n", "# Wrap as a callable so Crossfitting does NOT try to clone-and-refit it per fold.\n", "model_fn = lambda X_: rf.predict(X_)\n", "print(f\"RandomForestRegressor fitted on full data\")\n", "print(f\"Train R²: {rf.score(X_values, y_values):.3f}\")\n" ] }, { "cell_type": "markdown", "id": "18766952", "metadata": {}, "source": [ "## Fit Flow Model on All Data\n", "\n", "The flow is trained **once on the full dataset** before cross-fitting, analogous to how\n", "EOTExplainer computes its whitening map on all n observations. Cross-fitting is used\n", "only for UEIF variance estimation — not for flow training.\n", "\n", "**Why this matters for binary high-d data (d=832, n≈600):**\n", "\n", "- Training on all n samples gives the flow maximum statistical power to learn the\n", " correlation structure (d²/2 ≈ 345K pairwise correlations, n=600 samples).\n", "- Training per fold (n/2 ≈ 300) halves the effective sample size and produces a\n", " weaker disentanglement map, increasing H_avg toward identity.\n", "- Architecture: `hidden_dim=256, num_blocks=3` gives ~1.3M parameters — sufficient to\n", " represent the pairwise correlation structure (compare default 119K parameters with\n", " `hidden_dim=64, num_blocks=1`).\n", "\n", "**Z-space DFI is the primary output** for this dataset. X-space attribution via the\n", "flow Jacobian H = dX/dZ can be unreliable for binary data because H_avg ≈ I even for\n", "a well-trained flow (samples in opposite binary modes produce Jacobians with opposing\n", "off-diagonal signs that cancel in the average).\n" ] }, { "cell_type": "code", "execution_count": 5, "id": "68f7e54a", "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "Training flow model on full dataset...\n", "[FDFI][INFO] Training flow model...\n", "Training complete: 2000 steps, final loss=1.4006\n", "Flow training completed in 6023.9s\n", "Z_full shape: (611, 832)\n" ] } ], "source": [ "# Train the flow ONCE on all data — analogous to EOT fitting whitening on all n.\n", "# Cross-fitting will share this pre-trained flow across folds (fit_flow=False per fold).\n", "print(\"Training flow model on full dataset...\")\n", "flow_start = time.time()\n", "\n", "fdfi_estimator = FlowExplainer(\n", " model_fn,\n", " X_values,\n", " hidden_dim=256, # larger than default (64); needed for d=832\n", " num_blocks=3, # larger than default (1)\n", " num_steps=2000,\n", " fit_flow=True,\n", " random_state=0,\n", " verbose=\"final\",\n", " compute_diagnostics=False, # skip diagnostics here; run separately below\n", ")\n", "\n", "flow_results = fdfi_estimator(X_values, y=y_values)\n", "\n", "flow_train_time = time.time() - flow_start\n", "print(f\"Flow training completed in {flow_train_time:.1f}s\")\n", "print(f\"Z_full shape: {fdfi_estimator.Z_full.shape}\")\n" ] }, { "cell_type": "markdown", "id": "1b05d3e7", "metadata": {}, "source": [ "## Individual Feature Inference With conf_int()\n" ] }, { "cell_type": "code", "execution_count": 6, "id": "d90b66b5", "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "Results saved to results/flowmatching_case_study/sens50_flow_dfi_results.csv\n" ] } ], "source": [ "# Z-space is the primary target for FlowExplainer on binary high-d data.\n", "# var_floor_c=1e-3: FlowExplainer CPI UEIFs are nearly constant across samples\n", "# (nsamples=100 stabilises y_tilde_mean), so raw SEs are structurally small\n", "# (group-level range 1e-4 to 3e-3). The floor 1e-3/sqrt(n) ≈ 4e-5 is ~3x\n", "# below the smallest observed group SE, adding <5% to any z-score, while\n", "# guarding against numerical zeros at the individual-feature level.\n", "feature_names = X.columns.tolist()\n", "\n", "ci_Z = fdfi_estimator.conf_int(\n", " alpha=0.05,\n", " target=\"Z\",\n", " var_floor_c=1e-3,\n", " var_floor_method=\"fixed\",\n", " margin=0.0,\n", " margin_method=\"auto\",\n", " margin_quantile=0.95,\n", " alternative=\"two-sided\",\n", " verbose=False,\n", ")\n", "\n", "ci_X = fdfi_estimator.conf_int(\n", " alpha=0.05,\n", " target=\"X\",\n", " var_floor_c=1e-3,\n", " var_floor_method=\"fixed\",\n", " margin=0.0,\n", " margin_method=\"auto\",\n", " margin_quantile=0.95,\n", " alternative=\"two-sided\",\n", " verbose=False,\n", ")\n", "\n", "results = pd.DataFrame({\n", " \"feature\": feature_names,\n", " \"dfiz\": ci_Z[\"score\"],\n", " \"dfiz_se\": ci_Z[\"se\"],\n", " \"dfiz_zscore\": ci_Z[\"zscore\"],\n", " \"dfiz_p_value\": ci_Z[\"pvalue\"],\n", " \"dfiz_ci_lower\": ci_Z[\"ci_lower\"],\n", " \"dfiz_ci_upper\": ci_Z[\"ci_upper\"],\n", " \"dfiz_reject_null\": ci_Z[\"reject_null\"],\n", " \"dfiz_ranking\": ci_Z[\"ranking\"],\n", " \"dfi\": ci_X[\"score\"],\n", " \"dfi_se\": ci_X[\"se\"],\n", " \"dfi_zscore\": ci_X[\"zscore\"],\n", " \"dfi_p_value\": ci_X[\"pvalue\"],\n", " \"dfi_ci_lower\": ci_X[\"ci_lower\"],\n", " \"dfi_ci_upper\": ci_X[\"ci_upper\"],\n", " \"dfi_reject_null\": ci_X[\"reject_null\"],\n", " \"dfi_ranking\": ci_X[\"ranking\"],\n", "})\n", "\n", "OUT_DIR = Path(\"results\") / \"flowmatching_case_study\"\n", "OUT_DIR.mkdir(parents=True, exist_ok=True)\n", "results.to_csv(OUT_DIR / f\"{outcome}_flow_dfi_results.csv\", index=False)\n", "print(f\"Results saved to {OUT_DIR / f'{outcome}_flow_dfi_results.csv'}\")\n" ] }, { "cell_type": "markdown", "id": "7ca49175", "metadata": {}, "source": [ "## Feature-Level Summary\n" ] }, { "cell_type": "code", "execution_count": 7, "id": "5baebd48", "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "=== Z-space feature summary (primary for FlowExplainer on binary data) ===\n", "==============================================================================\n", "Feature Importance Results\n", "==============================================================================\n", "Method: FlowExplainer\n", "Number of units: 832\n", "Significance level: 0.05\n", "Alternative: two-sided\n", "Margin method: mixture\n", "Practical margin: 0.0019\n", "------------------------------------------------------------------------------\n", " Feature Estimate Std Err CI Lower CI Upper P-value Sig\n", "------------------------------------------------------------------------------\n", " 0 0.0002 0.0002 -0.0001 0.0006 0.0000 ***\n", " 1 0.0011 0.0010 -0.0009 0.0031 0.4037 \n", " 2 0.0001 0.0001 -0.0001 0.0002 0.0000 ***\n", " 3 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 4 0.0020 0.0020 -0.0019 0.0059 0.9724 \n", " 5 0.0012 0.0012 -0.0012 0.0037 0.5796 \n", " 6 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 7 0.0005 0.0005 -0.0004 0.0014 0.0014 ***\n", " 8 0.0009 0.0009 -0.0009 0.0027 0.2763 \n", " 9 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 10 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 11 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 12 0.0001 0.0001 -0.0001 0.0002 0.0000 ***\n", " 13 0.0003 0.0003 -0.0002 0.0008 0.0000 ***\n", " 14 0.0002 0.0001 -0.0001 0.0005 0.0000 ***\n", " 15 0.0001 0.0001 -0.0001 0.0002 0.0000 ***\n", " 16 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 17 0.0010 0.0010 -0.0010 0.0030 0.3866 \n", " 18 0.0002 0.0002 -0.0002 0.0006 0.0000 ***\n", " 19 0.0004 0.0001 0.0003 0.0006 0.0000 ***\n", " 20 0.0003 0.0002 -0.0002 0.0007 0.0000 ***\n", " 21 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 22 0.0001 0.0001 -0.0001 0.0002 0.0000 ***\n", " 23 0.0018 0.0018 -0.0018 0.0055 0.9694 \n", " 24 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 25 0.0002 0.0002 -0.0002 0.0007 0.0000 ***\n", " 26 0.0005 0.0005 -0.0005 0.0014 0.0020 ***\n", " 27 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 28 0.0001 0.0001 -0.0001 0.0003 0.0000 ***\n", " 29 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 30 0.0000 0.0001 -0.0001 0.0001 0.0000 ***\n", " 31 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 32 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 33 0.0003 0.0002 -0.0002 0.0007 0.0000 ***\n", " 34 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 35 0.0030 0.0030 -0.0028 0.0088 0.7235 \n", " 36 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 37 0.0001 0.0001 -0.0001 0.0003 0.0000 ***\n", " 38 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 39 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 40 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 41 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 42 0.0000 0.0001 -0.0001 0.0001 0.0000 ***\n", " 43 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 44 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 45 0.0001 0.0001 -0.0001 0.0002 0.0000 ***\n", " 46 0.0008 0.0008 -0.0007 0.0023 0.1376 \n", " 47 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 48 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 49 0.0001 0.0001 -0.0000 0.0002 0.0000 ***\n", " 50 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 51 0.0001 0.0001 -0.0001 0.0002 0.0000 ***\n", " 52 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 53 0.0000 0.0001 -0.0001 0.0002 0.0000 ***\n", " 54 0.0017 0.0017 -0.0017 0.0051 0.9173 \n", " 55 0.0014 0.0014 -0.0014 0.0042 0.7314 \n", " 56 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 57 0.0001 0.0001 -0.0001 0.0003 0.0000 ***\n", " 58 0.0001 0.0001 0.0000 0.0003 0.0000 ***\n", " 59 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 60 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 61 0.0001 0.0000 -0.0000 0.0002 0.0000 ***\n", " 62 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 63 0.0001 0.0001 -0.0001 0.0002 0.0000 ***\n", " 64 0.0017 0.0017 -0.0016 0.0050 0.8949 \n", " 65 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 66 0.0001 0.0001 -0.0001 0.0003 0.0000 ***\n", " 67 0.0006 0.0006 -0.0006 0.0018 0.0251 **\n", " 68 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 69 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 70 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 71 0.0017 0.0017 -0.0017 0.0051 0.9086 \n", " 72 0.0001 0.0001 -0.0001 0.0003 0.0000 ***\n", " 73 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 74 0.0000 0.0001 -0.0001 0.0002 0.0000 ***\n", " 75 0.0004 0.0004 -0.0004 0.0012 0.0001 ***\n", " 76 0.0017 0.0017 -0.0016 0.0050 0.8944 \n", " 77 0.0036 0.0016 0.0006 0.0067 0.2694 \n", " 78 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 79 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 80 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 81 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 82 0.0016 0.0016 -0.0015 0.0048 0.8454 \n", " 83 0.0003 0.0003 -0.0003 0.0010 0.0000 ***\n", " 84 0.0001 0.0001 -0.0001 0.0002 0.0000 ***\n", " 85 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 86 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 87 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 88 0.0014 0.0007 0.0001 0.0027 0.4358 \n", " 89 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 90 0.0020 0.0020 -0.0019 0.0058 0.9741 \n", " 91 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 92 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 93 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 94 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 95 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 96 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 97 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 98 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 99 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 100 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 101 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 102 0.0001 0.0001 -0.0001 0.0002 0.0000 ***\n", " 103 0.0000 0.0000 -0.0000 0.0001 0.0000 ***\n", " 104 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 105 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 106 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 107 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 108 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 109 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 110 0.0002 0.0003 -0.0002 0.0007 0.0000 ***\n", " 111 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 112 0.0001 0.0001 -0.0001 0.0004 0.0000 ***\n", " 113 0.0001 0.0001 -0.0001 0.0002 0.0000 ***\n", " 114 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 115 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 116 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 117 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 118 0.0001 0.0001 -0.0000 0.0002 0.0000 ***\n", " 119 0.0001 0.0001 -0.0001 0.0003 0.0000 ***\n", " 120 0.0001 0.0001 -0.0001 0.0003 0.0000 ***\n", " 121 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 122 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 123 0.0001 0.0001 -0.0000 0.0002 0.0000 ***\n", " 124 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 125 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 126 0.0000 0.0001 -0.0001 0.0002 0.0000 ***\n", " 127 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 128 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 129 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 130 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 131 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 132 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 133 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 134 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 135 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 136 0.0001 0.0001 -0.0001 0.0002 0.0000 ***\n", " 137 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 138 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 139 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 140 0.0006 0.0005 -0.0005 0.0016 0.0135 **\n", " 141 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 142 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 143 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 144 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 145 0.0010 0.0010 -0.0009 0.0028 0.3106 \n", " 146 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 147 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 148 0.0011 0.0011 -0.0010 0.0032 0.4464 \n", " 149 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 150 0.0001 0.0001 -0.0000 0.0003 0.0000 ***\n", " 151 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 152 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 153 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 154 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 155 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 156 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 157 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 158 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 159 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 160 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 161 0.0003 0.0001 0.0002 0.0005 0.0000 ***\n", " 162 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 163 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 164 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 165 0.0001 0.0001 -0.0001 0.0002 0.0000 ***\n", " 166 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 167 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 168 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 169 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 170 0.0007 0.0006 -0.0005 0.0019 0.0458 **\n", " 171 0.0000 0.0001 -0.0001 0.0002 0.0000 ***\n", " 172 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 173 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 174 0.0000 0.0001 -0.0001 0.0001 0.0000 ***\n", " 175 0.0001 0.0001 -0.0001 0.0002 0.0000 ***\n", " 176 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 177 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 178 0.0002 0.0002 -0.0002 0.0006 0.0000 ***\n", " 179 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 180 0.0001 0.0001 -0.0001 0.0003 0.0000 ***\n", " 181 0.0015 0.0015 -0.0014 0.0044 0.7778 \n", " 182 0.0011 0.0011 -0.0009 0.0032 0.4629 \n", " 183 0.0016 0.0016 -0.0015 0.0048 0.8677 \n", " 184 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 185 0.0001 0.0001 -0.0001 0.0002 0.0000 ***\n", " 186 0.0001 0.0001 -0.0001 0.0003 0.0000 ***\n", " 187 0.0014 0.0014 -0.0014 0.0042 0.7259 \n", " 188 0.0000 0.0001 -0.0001 0.0001 0.0000 ***\n", " 189 0.0001 0.0001 -0.0001 0.0003 0.0000 ***\n", " 190 0.0008 0.0008 -0.0007 0.0023 0.1345 \n", " 191 0.0001 0.0001 -0.0001 0.0004 0.0000 ***\n", " 192 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 193 0.0001 0.0000 -0.0000 0.0001 0.0000 ***\n", " 194 0.0011 0.0011 -0.0010 0.0032 0.4282 \n", " 195 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 196 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 197 0.0003 0.0003 -0.0002 0.0009 0.0000 ***\n", " 198 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 199 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 200 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 201 0.0005 0.0005 -0.0005 0.0014 0.0020 ***\n", " 202 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 203 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 204 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 205 0.0000 0.0000 -0.0000 0.0001 0.0000 ***\n", " 206 0.0001 0.0001 -0.0001 0.0004 0.0000 ***\n", " 207 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 208 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 209 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 210 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 211 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 212 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 213 0.0011 0.0011 -0.0010 0.0032 0.4525 \n", " 214 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 215 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 216 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 217 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 218 0.0002 0.0002 -0.0002 0.0007 0.0000 ***\n", " 219 0.0010 0.0010 -0.0010 0.0030 0.3814 \n", " 220 0.0001 0.0001 -0.0001 0.0003 0.0000 ***\n", " 221 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 222 0.0000 0.0001 -0.0001 0.0001 0.0000 ***\n", " 223 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 224 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 225 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 226 0.0001 0.0001 -0.0000 0.0002 0.0000 ***\n", " 227 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 228 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 229 0.0001 0.0001 -0.0001 0.0003 0.0000 ***\n", " 230 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 231 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 232 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 233 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 234 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 235 0.0012 0.0011 -0.0011 0.0034 0.4982 \n", " 236 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 237 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 238 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 239 0.0001 0.0001 -0.0001 0.0002 0.0000 ***\n", " 240 0.0001 0.0001 -0.0001 0.0003 0.0000 ***\n", " 241 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 242 0.0000 0.0001 -0.0001 0.0002 0.0000 ***\n", " 243 0.0001 0.0001 -0.0001 0.0003 0.0000 ***\n", " 244 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 245 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 246 0.0001 0.0001 -0.0001 0.0003 0.0000 ***\n", " 247 0.0001 0.0001 -0.0001 0.0003 0.0000 ***\n", " 248 0.0001 0.0001 -0.0000 0.0002 0.0000 ***\n", " 249 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 250 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 251 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 252 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 253 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 254 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 255 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 256 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 257 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 258 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 259 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 260 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 261 0.0001 0.0001 0.0000 0.0002 0.0000 ***\n", " 262 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 263 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 264 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 265 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 266 0.0000 0.0000 -0.0000 0.0001 0.0000 ***\n", " 267 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 268 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 269 0.0001 0.0001 -0.0001 0.0004 0.0000 ***\n", " 270 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 271 0.0001 0.0000 0.0000 0.0002 0.0000 ***\n", " 272 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 273 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 274 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 275 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 276 0.0001 0.0001 -0.0001 0.0002 0.0000 ***\n", " 277 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 278 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 279 0.0001 0.0001 -0.0001 0.0003 0.0000 ***\n", " 280 0.0016 0.0016 -0.0015 0.0047 0.8273 \n", " 281 0.0001 0.0001 -0.0001 0.0002 0.0000 ***\n", " 282 0.0004 0.0004 -0.0003 0.0011 0.0000 ***\n", " 283 0.0001 0.0000 -0.0000 0.0001 0.0000 ***\n", " 284 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 285 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 286 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 287 0.0008 0.0003 0.0002 0.0013 0.0000 ***\n", " 288 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 289 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 290 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 291 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 292 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 293 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 294 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 295 0.0001 0.0001 -0.0001 0.0004 0.0000 ***\n", " 296 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 297 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 298 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 299 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 300 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 301 0.0002 0.0001 -0.0001 0.0004 0.0000 ***\n", " 302 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 303 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 304 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 305 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 306 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 307 0.0002 0.0002 -0.0002 0.0005 0.0000 ***\n", " 308 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 309 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 310 0.0001 0.0001 -0.0001 0.0003 0.0000 ***\n", " 311 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 312 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 313 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 314 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 315 0.0000 0.0000 -0.0000 0.0001 0.0000 ***\n", " 316 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 317 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 318 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 319 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 320 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 321 0.0000 0.0001 -0.0001 0.0002 0.0000 ***\n", " 322 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 323 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 324 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 325 0.0001 0.0001 -0.0001 0.0004 0.0000 ***\n", " 326 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 327 0.0020 0.0020 -0.0019 0.0059 0.9637 \n", " 328 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 329 0.0000 0.0000 -0.0000 0.0001 0.0000 ***\n", " 330 0.0000 0.0001 -0.0001 0.0001 0.0000 ***\n", " 331 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 332 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 333 0.0001 0.0000 -0.0000 0.0001 0.0000 ***\n", " 334 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 335 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 336 0.0001 0.0001 -0.0001 0.0004 0.0000 ***\n", " 337 0.0001 0.0001 -0.0001 0.0002 0.0000 ***\n", " 338 0.0001 0.0001 -0.0001 0.0002 0.0000 ***\n", " 339 0.0000 0.0001 -0.0001 0.0002 0.0000 ***\n", " 340 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 341 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 342 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 343 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 344 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 345 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 346 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 347 0.0000 0.0000 -0.0000 0.0001 0.0000 ***\n", " 348 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 349 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 350 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 351 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 352 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 353 0.0002 0.0001 -0.0000 0.0003 0.0000 ***\n", " 354 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 355 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 356 0.0007 0.0007 -0.0007 0.0021 0.0937 *\n", " 357 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 358 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 359 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 360 0.0003 0.0003 -0.0003 0.0010 0.0000 ***\n", " 361 0.0007 0.0007 -0.0007 0.0022 0.1168 \n", " 362 0.0005 0.0003 -0.0002 0.0011 0.0000 ***\n", " 363 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 364 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 365 0.0002 0.0002 -0.0002 0.0006 0.0000 ***\n", " 366 0.0002 0.0001 0.0001 0.0003 0.0000 ***\n", " 367 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 368 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 369 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 370 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 371 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 372 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 373 0.0001 0.0001 -0.0001 0.0002 0.0000 ***\n", " 374 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 375 0.0001 0.0000 -0.0000 0.0002 0.0000 ***\n", " 376 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 377 0.0001 0.0001 -0.0001 0.0004 0.0000 ***\n", " 378 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 379 0.0003 0.0002 -0.0001 0.0007 0.0000 ***\n", " 380 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 381 0.0000 0.0001 -0.0001 0.0001 0.0000 ***\n", " 382 0.0017 0.0017 -0.0016 0.0049 0.8735 \n", " 383 0.0000 0.0001 -0.0001 0.0001 0.0000 ***\n", " 384 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 385 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 386 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 387 0.0001 0.0001 -0.0000 0.0002 0.0000 ***\n", " 388 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 389 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 390 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 391 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 392 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 393 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 394 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 395 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 396 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 397 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 398 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 399 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 400 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 401 0.0009 0.0009 -0.0008 0.0025 0.2172 \n", " 402 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 403 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 404 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 405 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 406 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 407 0.0020 0.0020 -0.0019 0.0058 0.9823 \n", " 408 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 409 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 410 0.0000 0.0001 -0.0001 0.0001 0.0000 ***\n", " 411 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 412 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 413 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 414 0.0001 0.0001 -0.0001 0.0002 0.0000 ***\n", " 415 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 416 0.0001 0.0001 -0.0001 0.0003 0.0000 ***\n", " 417 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 418 0.0001 0.0001 -0.0001 0.0004 0.0000 ***\n", " 419 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 420 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 421 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 422 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 423 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 424 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 425 0.0001 0.0001 -0.0000 0.0002 0.0000 ***\n", " 426 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 427 0.0001 0.0001 -0.0001 0.0003 0.0000 ***\n", " 428 0.0000 0.0001 -0.0001 0.0001 0.0000 ***\n", " 429 0.0003 0.0003 -0.0003 0.0009 0.0000 ***\n", " 430 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 431 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 432 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 433 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 434 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 435 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 436 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 437 0.0001 0.0001 -0.0000 0.0003 0.0000 ***\n", " 438 0.0000 0.0001 -0.0001 0.0002 0.0000 ***\n", " 439 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 440 0.0001 0.0001 -0.0001 0.0002 0.0000 ***\n", " 441 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 442 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 443 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 444 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 445 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 446 0.0001 0.0001 -0.0001 0.0002 0.0000 ***\n", " 447 0.0000 0.0001 -0.0001 0.0001 0.0000 ***\n", " 448 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 449 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 450 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 451 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 452 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 453 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 454 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 455 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 456 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 457 0.0001 0.0001 -0.0001 0.0002 0.0000 ***\n", " 458 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 459 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 460 0.0001 0.0001 -0.0000 0.0002 0.0000 ***\n", " 461 0.0001 0.0001 -0.0001 0.0002 0.0000 ***\n", " 462 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 463 0.0003 0.0002 0.0001 0.0006 0.0000 ***\n", " 464 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 465 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 466 0.0001 0.0001 -0.0001 0.0003 0.0000 ***\n", " 467 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 468 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 469 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 470 0.0001 0.0001 -0.0001 0.0003 0.0000 ***\n", " 471 0.0000 0.0001 -0.0001 0.0002 0.0000 ***\n", " 472 0.0001 0.0001 -0.0001 0.0002 0.0000 ***\n", " 473 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 474 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 475 0.0007 0.0007 -0.0006 0.0019 0.0533 *\n", " 476 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 477 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 478 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 479 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 480 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 481 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 482 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 483 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 484 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 485 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 486 0.0001 0.0001 -0.0001 0.0004 0.0000 ***\n", " 487 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 488 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 489 0.0001 0.0001 -0.0001 0.0002 0.0000 ***\n", " 490 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 491 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 492 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 493 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 494 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 495 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 496 0.0000 0.0001 -0.0001 0.0001 0.0000 ***\n", " 497 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 498 0.0001 0.0001 -0.0001 0.0004 0.0000 ***\n", " 499 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 500 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 501 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 502 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 503 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 504 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 505 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 506 0.0002 0.0002 -0.0001 0.0005 0.0000 ***\n", " 507 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 508 0.0001 0.0001 -0.0001 0.0003 0.0000 ***\n", " 509 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 510 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 511 0.0004 0.0004 -0.0003 0.0011 0.0000 ***\n", " 512 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 513 0.0001 0.0001 -0.0000 0.0002 0.0000 ***\n", " 514 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 515 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 516 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 517 0.0001 0.0000 -0.0000 0.0002 0.0000 ***\n", " 518 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 519 0.0001 0.0001 -0.0001 0.0004 0.0000 ***\n", " 520 0.0001 0.0001 -0.0001 0.0003 0.0000 ***\n", " 521 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 522 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 523 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 524 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 525 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 526 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 527 0.0001 0.0001 -0.0001 0.0002 0.0000 ***\n", " 528 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 529 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 530 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 531 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 532 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 533 0.0002 0.0002 -0.0002 0.0005 0.0000 ***\n", " 534 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 535 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 536 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 537 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 538 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 539 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 540 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 541 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 542 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 543 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 544 0.0001 0.0001 -0.0001 0.0002 0.0000 ***\n", " 545 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 546 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 547 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 548 0.0000 0.0000 -0.0000 0.0001 0.0000 ***\n", " 549 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 550 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 551 0.0001 0.0001 0.0000 0.0003 0.0000 ***\n", " 552 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 553 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 554 0.0001 0.0001 -0.0001 0.0003 0.0000 ***\n", " 555 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 556 0.0000 0.0001 -0.0001 0.0002 0.0000 ***\n", " 557 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 558 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 559 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 560 0.0002 0.0002 -0.0002 0.0006 0.0000 ***\n", " 561 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 562 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 563 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 564 0.0002 0.0002 -0.0002 0.0006 0.0000 ***\n", " 565 0.0001 0.0000 -0.0000 0.0002 0.0000 ***\n", " 566 0.0021 0.0020 -0.0018 0.0059 0.9418 \n", " 567 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 568 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 569 0.0006 0.0006 -0.0005 0.0017 0.0241 **\n", " 570 0.0016 0.0016 -0.0016 0.0049 0.8703 \n", " 571 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 572 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 573 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 574 0.0002 0.0001 -0.0001 0.0005 0.0000 ***\n", " 575 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 576 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 577 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 578 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 579 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 580 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 581 0.0002 0.0001 0.0001 0.0003 0.0000 ***\n", " 582 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 583 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 584 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 585 0.0002 0.0001 -0.0001 0.0004 0.0000 ***\n", " 586 0.0002 0.0002 -0.0002 0.0006 0.0000 ***\n", " 587 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 588 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 589 0.0000 0.0001 -0.0001 0.0002 0.0000 ***\n", " 590 0.0001 0.0000 -0.0000 0.0002 0.0000 ***\n", " 591 0.0001 0.0001 -0.0001 0.0004 0.0000 ***\n", " 592 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 593 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 594 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 595 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 596 0.0015 0.0015 -0.0015 0.0045 0.7922 \n", " 597 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 598 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 599 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 600 0.0000 0.0001 -0.0001 0.0001 0.0000 ***\n", " 601 0.0002 0.0002 -0.0002 0.0007 0.0000 ***\n", " 602 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 603 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 604 0.0001 0.0001 -0.0001 0.0003 0.0000 ***\n", " 605 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 606 0.0001 0.0001 -0.0001 0.0002 0.0000 ***\n", " 607 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 608 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 609 0.0000 0.0001 -0.0001 0.0001 0.0000 ***\n", " 610 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 611 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 612 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 613 0.0002 0.0002 -0.0002 0.0007 0.0000 ***\n", " 614 0.0002 0.0002 -0.0001 0.0006 0.0000 ***\n", " 615 0.0002 0.0001 -0.0000 0.0004 0.0000 ***\n", " 616 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 617 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 618 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 619 0.0000 0.0001 -0.0001 0.0001 0.0000 ***\n", " 620 0.0000 0.0000 -0.0000 0.0001 0.0000 ***\n", " 621 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 622 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 623 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 624 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 625 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 626 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 627 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 628 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 629 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 630 0.0001 0.0001 0.0000 0.0002 0.0000 ***\n", " 631 0.0000 0.0001 -0.0001 0.0002 0.0000 ***\n", " 632 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 633 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 634 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 635 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 636 0.0002 0.0001 -0.0001 0.0004 0.0000 ***\n", " 637 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 638 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 639 0.0002 0.0001 -0.0001 0.0004 0.0000 ***\n", " 640 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 641 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 642 0.0001 0.0001 -0.0001 0.0002 0.0000 ***\n", " 643 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 644 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 645 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 646 0.0001 0.0001 -0.0001 0.0004 0.0000 ***\n", " 647 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 648 0.0000 0.0000 -0.0000 0.0001 0.0000 ***\n", " 649 0.0001 0.0001 -0.0001 0.0002 0.0000 ***\n", " 650 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 651 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 652 0.0001 0.0001 -0.0001 0.0002 0.0000 ***\n", " 653 0.0001 0.0001 -0.0001 0.0002 0.0000 ***\n", " 654 0.0001 0.0001 -0.0001 0.0002 0.0000 ***\n", " 655 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 656 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 657 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 658 0.0003 0.0001 0.0001 0.0004 0.0000 ***\n", " 659 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 660 0.0001 0.0001 -0.0001 0.0004 0.0000 ***\n", " 661 0.0001 0.0001 -0.0001 0.0003 0.0000 ***\n", " 662 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 663 0.0003 0.0003 -0.0002 0.0008 0.0000 ***\n", " 664 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 665 0.0001 0.0001 -0.0000 0.0002 0.0000 ***\n", " 666 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 667 0.0007 0.0007 -0.0006 0.0021 0.0786 *\n", " 668 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 669 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 670 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 671 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 672 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 673 0.0001 0.0001 -0.0001 0.0002 0.0000 ***\n", " 674 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 675 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 676 0.0014 0.0014 -0.0013 0.0041 0.7044 \n", " 677 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 678 0.0002 0.0001 -0.0001 0.0004 0.0000 ***\n", " 679 0.0001 0.0001 0.0000 0.0002 0.0000 ***\n", " 680 0.0002 0.0002 -0.0002 0.0005 0.0000 ***\n", " 681 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 682 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 683 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 684 0.0001 0.0000 -0.0000 0.0002 0.0000 ***\n", " 685 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 686 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 687 0.0001 0.0001 -0.0001 0.0002 0.0000 ***\n", " 688 0.0006 0.0006 -0.0006 0.0018 0.0347 **\n", " 689 0.0012 0.0012 -0.0012 0.0036 0.5505 \n", " 690 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 691 0.0020 0.0020 -0.0019 0.0059 0.9725 \n", " 692 0.0020 0.0020 -0.0019 0.0059 0.9705 \n", " 693 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 694 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 695 0.0001 0.0001 -0.0001 0.0003 0.0000 ***\n", " 696 0.0001 0.0000 -0.0000 0.0002 0.0000 ***\n", " 697 0.0013 0.0013 -0.0013 0.0039 0.6555 \n", " 698 0.0000 0.0001 -0.0001 0.0001 0.0000 ***\n", " 699 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 700 0.0002 0.0002 -0.0002 0.0005 0.0000 ***\n", " 701 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 702 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 703 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 704 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 705 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 706 0.0003 0.0002 -0.0000 0.0006 0.0000 ***\n", " 707 0.0001 0.0001 -0.0001 0.0004 0.0000 ***\n", " 708 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 709 0.0002 0.0001 -0.0000 0.0005 0.0000 ***\n", " 710 0.0000 0.0001 -0.0001 0.0001 0.0000 ***\n", " 711 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 712 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 713 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 714 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 715 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 716 0.0001 0.0001 -0.0001 0.0002 0.0000 ***\n", " 717 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 718 0.0000 0.0001 -0.0001 0.0001 0.0000 ***\n", " 719 0.0001 0.0001 -0.0001 0.0003 0.0000 ***\n", " 720 0.0001 0.0001 -0.0001 0.0002 0.0000 ***\n", " 721 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 722 0.0006 0.0005 -0.0004 0.0016 0.0114 **\n", " 723 0.0006 0.0006 -0.0006 0.0018 0.0378 **\n", " 724 0.0002 0.0002 -0.0002 0.0005 0.0000 ***\n", " 725 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 726 0.0001 0.0001 -0.0001 0.0003 0.0000 ***\n", " 727 0.0012 0.0012 -0.0012 0.0036 0.5765 \n", " 728 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 729 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 730 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 731 0.0000 0.0000 -0.0000 0.0001 0.0000 ***\n", " 732 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 733 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 734 0.0000 0.0000 -0.0000 0.0001 0.0000 ***\n", " 735 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 736 0.0000 0.0000 -0.0000 0.0001 0.0000 ***\n", " 737 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 738 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 739 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 740 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 741 0.0006 0.0006 -0.0006 0.0019 0.0387 **\n", " 742 0.0001 0.0001 -0.0001 0.0003 0.0000 ***\n", " 743 0.0000 0.0001 -0.0001 0.0002 0.0000 ***\n", " 744 0.0010 0.0010 -0.0010 0.0030 0.3621 \n", " 745 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 746 0.0016 0.0016 -0.0015 0.0048 0.8463 \n", " 747 0.0002 0.0002 -0.0002 0.0006 0.0000 ***\n", " 748 0.0001 0.0001 -0.0001 0.0002 0.0000 ***\n", " 749 0.0001 0.0001 0.0000 0.0003 0.0000 ***\n", " 750 0.0001 0.0001 -0.0001 0.0002 0.0000 ***\n", " 751 0.0013 0.0013 -0.0013 0.0039 0.6373 \n", " 752 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 753 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 754 0.0000 0.0001 -0.0001 0.0002 0.0000 ***\n", " 755 0.0002 0.0002 -0.0002 0.0007 0.0000 ***\n", " 756 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 757 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 758 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 759 0.0020 0.0020 -0.0019 0.0060 0.9570 \n", " 760 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 761 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 762 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 763 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 764 0.0000 0.0000 -0.0000 0.0001 0.0000 ***\n", " 765 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 766 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 767 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 768 0.0003 0.0003 -0.0003 0.0009 0.0000 ***\n", " 769 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 770 0.0001 0.0001 -0.0001 0.0002 0.0000 ***\n", " 771 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 772 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 773 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 774 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 775 0.0001 0.0001 -0.0000 0.0003 0.0000 ***\n", " 776 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 777 0.0001 0.0001 -0.0000 0.0002 0.0000 ***\n", " 778 0.0008 0.0008 -0.0008 0.0024 0.1767 \n", " 779 0.0001 0.0000 -0.0000 0.0001 0.0000 ***\n", " 780 0.0001 0.0001 -0.0001 0.0002 0.0000 ***\n", " 781 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 782 0.0001 0.0001 -0.0001 0.0004 0.0000 ***\n", " 783 0.0014 0.0014 -0.0013 0.0040 0.6900 \n", " 784 0.0001 0.0001 -0.0001 0.0002 0.0000 ***\n", " 785 0.0002 0.0002 -0.0002 0.0006 0.0000 ***\n", " 786 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 787 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 788 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 789 0.0000 0.0001 -0.0001 0.0002 0.0000 ***\n", " 790 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 791 0.0001 0.0001 -0.0001 0.0002 0.0000 ***\n", " 792 0.0001 0.0001 -0.0001 0.0004 0.0000 ***\n", " 793 0.0002 0.0001 -0.0001 0.0004 0.0000 ***\n", " 794 0.0003 0.0003 -0.0002 0.0008 0.0000 ***\n", " 795 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 796 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 797 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 798 0.0002 0.0002 -0.0002 0.0005 0.0000 ***\n", " 799 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 800 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 801 0.0001 0.0001 -0.0001 0.0004 0.0000 ***\n", " 802 0.0002 0.0001 0.0001 0.0003 0.0000 ***\n", " 803 0.0001 0.0001 -0.0000 0.0002 0.0000 ***\n", " 804 0.0016 0.0016 -0.0016 0.0049 0.8676 \n", " 805 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 806 0.0029 0.0003 0.0022 0.0036 0.0038 ***\n", " 807 0.0005 0.0001 0.0003 0.0007 0.0000 ***\n", " 808 0.0001 0.0001 -0.0001 0.0003 0.0000 ***\n", " 809 0.0001 0.0001 -0.0001 0.0002 0.0000 ***\n", " 810 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 811 0.0000 0.0000 -0.0000 0.0001 0.0000 ***\n", " 812 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 813 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 814 0.0001 0.0001 -0.0001 0.0002 0.0000 ***\n", " 815 0.0016 0.0002 0.0011 0.0021 0.1609 \n", " 816 0.0010 0.0003 0.0004 0.0016 0.0011 ***\n", " 817 0.0001 0.0001 -0.0001 0.0003 0.0000 ***\n", " 818 0.0001 0.0001 -0.0001 0.0003 0.0000 ***\n", " 819 0.0000 0.0000 -0.0000 0.0001 0.0000 ***\n", " 820 0.0000 0.0000 -0.0000 0.0001 0.0000 ***\n", " 821 0.0004 0.0001 0.0003 0.0006 0.0000 ***\n", " 822 0.0005 0.0005 -0.0005 0.0015 0.0072 ***\n", " 823 0.0002 0.0002 -0.0001 0.0005 0.0000 ***\n", " 824 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 825 0.0002 0.0001 -0.0001 0.0005 0.0000 ***\n", " 826 0.0004 0.0001 0.0002 0.0006 0.0000 ***\n", " 827 0.0003 0.0001 0.0001 0.0005 0.0000 ***\n", " 828 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 829 0.0002 0.0002 -0.0001 0.0005 0.0000 ***\n", " 830 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 831 0.0001 0.0000 0.0000 0.0002 0.0000 ***\n", "==============================================================================\n", "Significant units: 778 / 832\n", "---\n", "Signif. codes: 0 '***' 0.01 '**' 0.05 '*' 0.1 ' ' 1\n", "==============================================================================\n", "\n", "=== X-space feature summary (secondary — Jacobian-based, may be noisy) ===\n", "==============================================================================\n", "Feature Importance Results\n", "==============================================================================\n", "Method: FlowExplainer\n", "Number of units: 832\n", "Significance level: 0.05\n", "Alternative: two-sided\n", "Margin method: mixture\n", "Practical margin: 0.0019\n", "------------------------------------------------------------------------------\n", " Feature Estimate Std Err CI Lower CI Upper P-value Sig\n", "------------------------------------------------------------------------------\n", " 0 0.0002 0.0002 -0.0001 0.0006 0.0000 ***\n", " 1 0.0011 0.0010 -0.0009 0.0031 0.4307 \n", " 2 0.0001 0.0001 -0.0000 0.0002 0.0000 ***\n", " 3 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 4 0.0016 0.0016 -0.0016 0.0049 0.8909 \n", " 5 0.0011 0.0011 -0.0010 0.0031 0.4424 \n", " 6 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 7 0.0004 0.0004 -0.0004 0.0012 0.0002 ***\n", " 8 0.0008 0.0008 -0.0008 0.0025 0.2213 \n", " 9 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 10 0.0000 0.0000 -0.0000 0.0001 0.0000 ***\n", " 11 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 12 0.0001 0.0001 -0.0001 0.0002 0.0000 ***\n", " 13 0.0002 0.0002 -0.0002 0.0006 0.0000 ***\n", " 14 0.0002 0.0001 -0.0000 0.0004 0.0000 ***\n", " 15 0.0001 0.0001 -0.0001 0.0002 0.0000 ***\n", " 16 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 17 0.0010 0.0010 -0.0009 0.0029 0.3814 \n", " 18 0.0002 0.0002 -0.0002 0.0006 0.0000 ***\n", " 19 0.0004 0.0001 0.0003 0.0006 0.0000 ***\n", " 20 0.0003 0.0002 -0.0001 0.0006 0.0000 ***\n", " 21 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 22 0.0001 0.0001 -0.0001 0.0002 0.0000 ***\n", " 23 0.0013 0.0013 -0.0012 0.0038 0.6440 \n", " 24 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 25 0.0002 0.0002 -0.0002 0.0006 0.0000 ***\n", " 26 0.0004 0.0004 -0.0004 0.0011 0.0001 ***\n", " 27 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 28 0.0001 0.0001 -0.0001 0.0004 0.0000 ***\n", " 29 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 30 0.0000 0.0001 -0.0001 0.0002 0.0000 ***\n", " 31 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 32 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 33 0.0003 0.0002 -0.0002 0.0007 0.0000 ***\n", " 34 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 35 0.0022 0.0022 -0.0021 0.0066 0.8732 \n", " 36 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 37 0.0001 0.0001 -0.0001 0.0003 0.0000 ***\n", " 38 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 39 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 40 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 41 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 42 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 43 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 44 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 45 0.0001 0.0001 -0.0001 0.0002 0.0000 ***\n", " 46 0.0008 0.0008 -0.0007 0.0023 0.1582 \n", " 47 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 48 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 49 0.0001 0.0001 0.0000 0.0002 0.0000 ***\n", " 50 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 51 0.0001 0.0001 -0.0001 0.0002 0.0000 ***\n", " 52 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 53 0.0001 0.0001 -0.0001 0.0002 0.0000 ***\n", " 54 0.0013 0.0013 -0.0012 0.0038 0.6488 \n", " 55 0.0012 0.0012 -0.0011 0.0035 0.5513 \n", " 56 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 57 0.0001 0.0001 -0.0001 0.0003 0.0000 ***\n", " 58 0.0002 0.0001 0.0000 0.0003 0.0000 ***\n", " 59 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 60 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 61 0.0001 0.0000 -0.0000 0.0002 0.0000 ***\n", " 62 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 63 0.0001 0.0001 -0.0000 0.0002 0.0000 ***\n", " 64 0.0013 0.0013 -0.0013 0.0040 0.6914 \n", " 65 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 66 0.0001 0.0001 -0.0001 0.0002 0.0000 ***\n", " 67 0.0005 0.0005 -0.0005 0.0015 0.0089 ***\n", " 68 0.0000 0.0000 -0.0000 0.0001 0.0000 ***\n", " 69 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 70 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 71 0.0013 0.0013 -0.0013 0.0039 0.6809 \n", " 72 0.0001 0.0001 -0.0001 0.0003 0.0000 ***\n", " 73 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 74 0.0001 0.0001 -0.0001 0.0002 0.0000 ***\n", " 75 0.0003 0.0003 -0.0003 0.0009 0.0000 ***\n", " 76 0.0012 0.0012 -0.0011 0.0035 0.5721 \n", " 77 0.0032 0.0014 0.0005 0.0059 0.3291 \n", " 78 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 79 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 80 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 81 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 82 0.0010 0.0010 -0.0009 0.0029 0.3494 \n", " 83 0.0003 0.0003 -0.0003 0.0008 0.0000 ***\n", " 84 0.0001 0.0001 -0.0001 0.0002 0.0000 ***\n", " 85 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 86 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 87 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 88 0.0013 0.0006 0.0001 0.0025 0.3433 \n", " 89 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 90 0.0016 0.0016 -0.0015 0.0047 0.8584 \n", " 91 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 92 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 93 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 94 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 95 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 96 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 97 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 98 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 99 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 100 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 101 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 102 0.0001 0.0001 -0.0001 0.0003 0.0000 ***\n", " 103 0.0000 0.0000 -0.0000 0.0001 0.0000 ***\n", " 104 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 105 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 106 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 107 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 108 0.0000 0.0000 -0.0000 0.0001 0.0000 ***\n", " 109 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 110 0.0002 0.0002 -0.0002 0.0007 0.0000 ***\n", " 111 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 112 0.0001 0.0001 -0.0001 0.0004 0.0000 ***\n", " 113 0.0001 0.0001 -0.0001 0.0002 0.0000 ***\n", " 114 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 115 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 116 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 117 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 118 0.0001 0.0001 0.0000 0.0002 0.0000 ***\n", " 119 0.0001 0.0001 -0.0001 0.0003 0.0000 ***\n", " 120 0.0001 0.0001 -0.0001 0.0003 0.0000 ***\n", " 121 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 122 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 123 0.0001 0.0001 -0.0000 0.0002 0.0000 ***\n", " 124 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 125 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 126 0.0000 0.0001 -0.0001 0.0002 0.0000 ***\n", " 127 0.0000 0.0000 -0.0000 0.0001 0.0000 ***\n", " 128 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 129 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 130 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 131 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 132 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 133 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 134 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 135 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 136 0.0001 0.0001 -0.0000 0.0002 0.0000 ***\n", " 137 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 138 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 139 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 140 0.0004 0.0004 -0.0004 0.0012 0.0005 ***\n", " 141 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 142 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 143 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 144 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 145 0.0008 0.0008 -0.0007 0.0022 0.1424 \n", " 146 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 147 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 148 0.0007 0.0007 -0.0007 0.0020 0.0885 *\n", " 149 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 150 0.0001 0.0001 -0.0000 0.0003 0.0000 ***\n", " 151 0.0001 0.0000 -0.0000 0.0001 0.0000 ***\n", " 152 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 153 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 154 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 155 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 156 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 157 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 158 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 159 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 160 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 161 0.0003 0.0001 0.0002 0.0005 0.0000 ***\n", " 162 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 163 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 164 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 165 0.0001 0.0001 -0.0001 0.0002 0.0000 ***\n", " 166 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 167 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 168 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 169 0.0001 0.0000 -0.0000 0.0001 0.0000 ***\n", " 170 0.0006 0.0005 -0.0005 0.0017 0.0215 **\n", " 171 0.0001 0.0001 -0.0001 0.0002 0.0000 ***\n", " 172 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 173 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 174 0.0001 0.0001 -0.0001 0.0002 0.0000 ***\n", " 175 0.0001 0.0001 -0.0001 0.0003 0.0000 ***\n", " 176 0.0000 0.0000 -0.0000 0.0001 0.0000 ***\n", " 177 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 178 0.0002 0.0002 -0.0002 0.0005 0.0000 ***\n", " 179 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 180 0.0001 0.0001 -0.0001 0.0003 0.0000 ***\n", " 181 0.0012 0.0012 -0.0011 0.0035 0.5738 \n", " 182 0.0010 0.0009 -0.0008 0.0027 0.3046 \n", " 183 0.0014 0.0014 -0.0013 0.0041 0.7384 \n", " 184 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 185 0.0001 0.0001 -0.0001 0.0002 0.0000 ***\n", " 186 0.0001 0.0001 -0.0001 0.0003 0.0000 ***\n", " 187 0.0011 0.0011 -0.0010 0.0032 0.4549 \n", " 188 0.0000 0.0001 -0.0001 0.0001 0.0000 ***\n", " 189 0.0001 0.0001 -0.0001 0.0003 0.0000 ***\n", " 190 0.0006 0.0006 -0.0006 0.0017 0.0268 **\n", " 191 0.0001 0.0001 -0.0001 0.0004 0.0000 ***\n", " 192 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 193 0.0001 0.0000 -0.0000 0.0001 0.0000 ***\n", " 194 0.0009 0.0009 -0.0008 0.0026 0.2462 \n", " 195 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 196 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 197 0.0003 0.0003 -0.0002 0.0009 0.0000 ***\n", " 198 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 199 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 200 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 201 0.0004 0.0004 -0.0004 0.0011 0.0001 ***\n", " 202 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 203 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 204 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 205 0.0001 0.0000 -0.0000 0.0002 0.0000 ***\n", " 206 0.0001 0.0001 -0.0001 0.0004 0.0000 ***\n", " 207 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 208 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 209 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 210 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 211 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 212 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 213 0.0009 0.0009 -0.0009 0.0027 0.3103 \n", " 214 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 215 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 216 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 217 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 218 0.0002 0.0002 -0.0001 0.0005 0.0000 ***\n", " 219 0.0008 0.0008 -0.0008 0.0024 0.1927 \n", " 220 0.0001 0.0001 -0.0001 0.0003 0.0000 ***\n", " 221 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 222 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 223 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 224 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 225 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 226 0.0001 0.0001 -0.0000 0.0002 0.0000 ***\n", " 227 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 228 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 229 0.0001 0.0001 -0.0001 0.0003 0.0000 ***\n", " 230 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 231 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 232 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 233 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 234 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 235 0.0010 0.0009 -0.0009 0.0028 0.3414 \n", " 236 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 237 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 238 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 239 0.0001 0.0001 -0.0001 0.0002 0.0000 ***\n", " 240 0.0001 0.0001 -0.0001 0.0003 0.0000 ***\n", " 241 0.0000 0.0000 -0.0000 0.0001 0.0000 ***\n", " 242 0.0001 0.0001 -0.0001 0.0002 0.0000 ***\n", " 243 0.0001 0.0001 -0.0001 0.0003 0.0000 ***\n", " 244 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 245 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 246 0.0001 0.0001 -0.0001 0.0003 0.0000 ***\n", " 247 0.0001 0.0001 -0.0001 0.0003 0.0000 ***\n", " 248 0.0001 0.0000 -0.0000 0.0002 0.0000 ***\n", " 249 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 250 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 251 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 252 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 253 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 254 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 255 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 256 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 257 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 258 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 259 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 260 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 261 0.0001 0.0001 0.0000 0.0002 0.0000 ***\n", " 262 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 263 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 264 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 265 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 266 0.0000 0.0000 -0.0000 0.0001 0.0000 ***\n", " 267 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 268 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 269 0.0001 0.0001 -0.0001 0.0004 0.0000 ***\n", " 270 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 271 0.0001 0.0000 0.0000 0.0002 0.0000 ***\n", " 272 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 273 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 274 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 275 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 276 0.0001 0.0001 -0.0001 0.0002 0.0000 ***\n", " 277 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 278 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 279 0.0001 0.0001 -0.0001 0.0003 0.0000 ***\n", " 280 0.0013 0.0012 -0.0012 0.0037 0.6202 \n", " 281 0.0001 0.0001 -0.0001 0.0002 0.0000 ***\n", " 282 0.0003 0.0003 -0.0003 0.0009 0.0000 ***\n", " 283 0.0001 0.0000 -0.0000 0.0002 0.0000 ***\n", " 284 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 285 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 286 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 287 0.0008 0.0003 0.0002 0.0013 0.0001 ***\n", " 288 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 289 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 290 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 291 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 292 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 293 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 294 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 295 0.0002 0.0001 -0.0001 0.0004 0.0000 ***\n", " 296 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 297 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 298 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 299 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 300 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 301 0.0002 0.0001 -0.0001 0.0004 0.0000 ***\n", " 302 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 303 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 304 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 305 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 306 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 307 0.0002 0.0002 -0.0001 0.0005 0.0000 ***\n", " 308 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 309 0.0000 0.0000 -0.0000 0.0001 0.0000 ***\n", " 310 0.0001 0.0001 -0.0001 0.0003 0.0000 ***\n", " 311 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 312 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 313 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 314 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 315 0.0000 0.0000 -0.0000 0.0001 0.0000 ***\n", " 316 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 317 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 318 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 319 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 320 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 321 0.0001 0.0001 -0.0000 0.0002 0.0000 ***\n", " 322 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 323 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 324 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 325 0.0001 0.0001 -0.0001 0.0003 0.0000 ***\n", " 326 0.0000 0.0000 -0.0000 0.0001 0.0000 ***\n", " 327 0.0016 0.0016 -0.0015 0.0048 0.8768 \n", " 328 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 329 0.0000 0.0000 -0.0000 0.0001 0.0000 ***\n", " 330 0.0001 0.0001 -0.0001 0.0002 0.0000 ***\n", " 331 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 332 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 333 0.0001 0.0000 -0.0000 0.0002 0.0000 ***\n", " 334 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 335 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 336 0.0001 0.0001 -0.0001 0.0004 0.0000 ***\n", " 337 0.0001 0.0001 -0.0001 0.0002 0.0000 ***\n", " 338 0.0001 0.0001 -0.0001 0.0002 0.0000 ***\n", " 339 0.0001 0.0001 -0.0001 0.0002 0.0000 ***\n", " 340 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 341 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 342 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 343 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 344 0.0000 0.0001 -0.0001 0.0001 0.0000 ***\n", " 345 0.0000 0.0000 -0.0000 0.0001 0.0000 ***\n", " 346 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 347 0.0000 0.0000 -0.0000 0.0001 0.0000 ***\n", " 348 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 349 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 350 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 351 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 352 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 353 0.0002 0.0001 -0.0000 0.0003 0.0000 ***\n", " 354 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 355 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 356 0.0007 0.0007 -0.0006 0.0020 0.0684 *\n", " 357 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 358 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 359 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 360 0.0003 0.0003 -0.0003 0.0010 0.0000 ***\n", " 361 0.0007 0.0007 -0.0006 0.0020 0.0780 *\n", " 362 0.0004 0.0003 -0.0001 0.0009 0.0000 ***\n", " 363 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 364 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 365 0.0002 0.0002 -0.0002 0.0006 0.0000 ***\n", " 366 0.0002 0.0001 0.0001 0.0004 0.0000 ***\n", " 367 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 368 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 369 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 370 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 371 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 372 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 373 0.0001 0.0001 -0.0001 0.0002 0.0000 ***\n", " 374 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 375 0.0001 0.0000 -0.0000 0.0002 0.0000 ***\n", " 376 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 377 0.0001 0.0001 -0.0001 0.0003 0.0000 ***\n", " 378 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 379 0.0002 0.0002 -0.0001 0.0006 0.0000 ***\n", " 380 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 381 0.0001 0.0001 -0.0001 0.0002 0.0000 ***\n", " 382 0.0015 0.0015 -0.0014 0.0043 0.7779 \n", " 383 0.0000 0.0001 -0.0001 0.0002 0.0000 ***\n", " 384 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 385 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 386 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 387 0.0001 0.0001 -0.0000 0.0002 0.0000 ***\n", " 388 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 389 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 390 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 391 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 392 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 393 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 394 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 395 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 396 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 397 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 398 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 399 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 400 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 401 0.0008 0.0008 -0.0008 0.0024 0.1973 \n", " 402 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 403 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 404 0.0000 0.0000 -0.0000 0.0001 0.0000 ***\n", " 405 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 406 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 407 0.0012 0.0012 -0.0011 0.0035 0.5754 \n", " 408 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 409 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 410 0.0000 0.0001 -0.0001 0.0001 0.0000 ***\n", " 411 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 412 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 413 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 414 0.0001 0.0001 -0.0001 0.0002 0.0000 ***\n", " 415 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 416 0.0001 0.0001 -0.0001 0.0003 0.0000 ***\n", " 417 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 418 0.0001 0.0001 -0.0001 0.0004 0.0000 ***\n", " 419 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 420 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 421 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 422 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 423 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 424 0.0000 0.0000 -0.0000 0.0001 0.0000 ***\n", " 425 0.0001 0.0000 -0.0000 0.0002 0.0000 ***\n", " 426 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 427 0.0001 0.0001 -0.0001 0.0003 0.0000 ***\n", " 428 0.0000 0.0001 -0.0001 0.0002 0.0000 ***\n", " 429 0.0002 0.0002 -0.0002 0.0007 0.0000 ***\n", " 430 0.0001 0.0001 -0.0001 0.0002 0.0000 ***\n", " 431 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 432 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 433 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 434 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 435 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 436 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 437 0.0001 0.0001 -0.0000 0.0003 0.0000 ***\n", " 438 0.0000 0.0001 -0.0001 0.0002 0.0000 ***\n", " 439 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 440 0.0001 0.0001 -0.0000 0.0002 0.0000 ***\n", " 441 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 442 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 443 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 444 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 445 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 446 0.0001 0.0001 -0.0001 0.0002 0.0000 ***\n", " 447 0.0000 0.0001 -0.0001 0.0002 0.0000 ***\n", " 448 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 449 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 450 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 451 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 452 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 453 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 454 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 455 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 456 0.0000 0.0000 -0.0000 0.0001 0.0000 ***\n", " 457 0.0001 0.0001 -0.0001 0.0002 0.0000 ***\n", " 458 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 459 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 460 0.0001 0.0001 -0.0000 0.0002 0.0000 ***\n", " 461 0.0001 0.0001 -0.0001 0.0002 0.0000 ***\n", " 462 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 463 0.0003 0.0001 0.0001 0.0006 0.0000 ***\n", " 464 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 465 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 466 0.0001 0.0001 -0.0001 0.0003 0.0000 ***\n", " 467 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 468 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 469 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 470 0.0001 0.0001 -0.0001 0.0003 0.0000 ***\n", " 471 0.0001 0.0001 -0.0001 0.0002 0.0000 ***\n", " 472 0.0001 0.0001 -0.0001 0.0002 0.0000 ***\n", " 473 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 474 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 475 0.0005 0.0005 -0.0005 0.0015 0.0097 ***\n", " 476 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 477 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 478 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 479 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 480 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 481 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 482 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 483 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 484 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 485 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 486 0.0001 0.0001 -0.0001 0.0004 0.0000 ***\n", " 487 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 488 0.0000 0.0000 -0.0000 0.0001 0.0000 ***\n", " 489 0.0001 0.0001 -0.0001 0.0002 0.0000 ***\n", " 490 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 491 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 492 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 493 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 494 0.0000 0.0000 -0.0000 0.0001 0.0000 ***\n", " 495 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 496 0.0001 0.0001 -0.0000 0.0002 0.0000 ***\n", " 497 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 498 0.0001 0.0001 -0.0001 0.0003 0.0000 ***\n", " 499 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 500 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 501 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 502 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 503 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 504 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 505 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 506 0.0002 0.0001 -0.0001 0.0004 0.0000 ***\n", " 507 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 508 0.0001 0.0001 -0.0001 0.0003 0.0000 ***\n", " 509 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 510 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 511 0.0003 0.0003 -0.0003 0.0010 0.0000 ***\n", " 512 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 513 0.0001 0.0001 -0.0000 0.0002 0.0000 ***\n", " 514 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 515 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 516 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 517 0.0001 0.0000 -0.0000 0.0002 0.0000 ***\n", " 518 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 519 0.0001 0.0001 -0.0001 0.0004 0.0000 ***\n", " 520 0.0001 0.0001 -0.0001 0.0003 0.0000 ***\n", " 521 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 522 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 523 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 524 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 525 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 526 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 527 0.0001 0.0001 -0.0001 0.0002 0.0000 ***\n", " 528 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 529 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 530 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 531 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 532 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 533 0.0002 0.0002 -0.0001 0.0005 0.0000 ***\n", " 534 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 535 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 536 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 537 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 538 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 539 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 540 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 541 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 542 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 543 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 544 0.0001 0.0001 -0.0001 0.0002 0.0000 ***\n", " 545 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 546 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 547 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 548 0.0001 0.0000 -0.0000 0.0001 0.0000 ***\n", " 549 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 550 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 551 0.0001 0.0001 0.0000 0.0003 0.0000 ***\n", " 552 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 553 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 554 0.0001 0.0001 -0.0001 0.0003 0.0000 ***\n", " 555 0.0000 0.0000 -0.0000 0.0001 0.0000 ***\n", " 556 0.0001 0.0001 -0.0001 0.0002 0.0000 ***\n", " 557 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 558 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 559 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 560 0.0002 0.0002 -0.0002 0.0006 0.0000 ***\n", " 561 0.0000 0.0000 -0.0000 0.0001 0.0000 ***\n", " 562 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 563 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 564 0.0002 0.0002 -0.0002 0.0006 0.0000 ***\n", " 565 0.0001 0.0001 -0.0000 0.0002 0.0000 ***\n", " 566 0.0017 0.0016 -0.0014 0.0048 0.9078 \n", " 567 0.0000 0.0000 -0.0000 0.0001 0.0000 ***\n", " 568 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 569 0.0006 0.0005 -0.0005 0.0016 0.0165 **\n", " 570 0.0012 0.0012 -0.0012 0.0036 0.5838 \n", " 571 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 572 0.0000 0.0000 -0.0000 0.0001 0.0000 ***\n", " 573 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 574 0.0002 0.0001 -0.0001 0.0005 0.0000 ***\n", " 575 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 576 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 577 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 578 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 579 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 580 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 581 0.0002 0.0001 0.0001 0.0003 0.0000 ***\n", " 582 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 583 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 584 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 585 0.0002 0.0001 -0.0001 0.0004 0.0000 ***\n", " 586 0.0002 0.0002 -0.0002 0.0006 0.0000 ***\n", " 587 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 588 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 589 0.0001 0.0001 -0.0001 0.0002 0.0000 ***\n", " 590 0.0001 0.0000 -0.0000 0.0002 0.0000 ***\n", " 591 0.0001 0.0001 -0.0001 0.0004 0.0000 ***\n", " 592 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 593 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 594 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 595 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 596 0.0014 0.0014 -0.0013 0.0041 0.7165 \n", " 597 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 598 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 599 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 600 0.0001 0.0001 -0.0001 0.0002 0.0000 ***\n", " 601 0.0002 0.0002 -0.0002 0.0007 0.0000 ***\n", " 602 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 603 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 604 0.0001 0.0001 -0.0001 0.0003 0.0000 ***\n", " 605 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 606 0.0001 0.0001 -0.0001 0.0002 0.0000 ***\n", " 607 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 608 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 609 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 610 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 611 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 612 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 613 0.0002 0.0002 -0.0002 0.0006 0.0000 ***\n", " 614 0.0002 0.0002 -0.0001 0.0005 0.0000 ***\n", " 615 0.0002 0.0001 -0.0000 0.0004 0.0000 ***\n", " 616 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 617 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 618 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 619 0.0000 0.0001 -0.0001 0.0002 0.0000 ***\n", " 620 0.0000 0.0000 -0.0000 0.0001 0.0000 ***\n", " 621 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 622 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 623 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 624 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 625 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 626 0.0000 0.0000 -0.0000 0.0001 0.0000 ***\n", " 627 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 628 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 629 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 630 0.0001 0.0001 0.0000 0.0002 0.0000 ***\n", " 631 0.0001 0.0001 -0.0001 0.0002 0.0000 ***\n", " 632 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 633 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 634 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 635 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 636 0.0001 0.0001 -0.0001 0.0003 0.0000 ***\n", " 637 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 638 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 639 0.0002 0.0001 -0.0001 0.0004 0.0000 ***\n", " 640 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 641 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 642 0.0001 0.0001 -0.0001 0.0002 0.0000 ***\n", " 643 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 644 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 645 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 646 0.0001 0.0001 -0.0001 0.0004 0.0000 ***\n", " 647 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 648 0.0001 0.0000 -0.0000 0.0001 0.0000 ***\n", " 649 0.0001 0.0001 -0.0001 0.0002 0.0000 ***\n", " 650 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 651 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 652 0.0001 0.0001 -0.0001 0.0002 0.0000 ***\n", " 653 0.0001 0.0001 -0.0000 0.0002 0.0000 ***\n", " 654 0.0001 0.0001 -0.0001 0.0002 0.0000 ***\n", " 655 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 656 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 657 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 658 0.0003 0.0001 0.0001 0.0004 0.0000 ***\n", " 659 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 660 0.0001 0.0001 -0.0001 0.0004 0.0000 ***\n", " 661 0.0001 0.0001 -0.0001 0.0003 0.0000 ***\n", " 662 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 663 0.0002 0.0002 -0.0002 0.0007 0.0000 ***\n", " 664 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 665 0.0001 0.0001 -0.0000 0.0002 0.0000 ***\n", " 666 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 667 0.0006 0.0006 -0.0005 0.0018 0.0341 **\n", " 668 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 669 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 670 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 671 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 672 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 673 0.0001 0.0001 -0.0001 0.0003 0.0000 ***\n", " 674 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 675 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 676 0.0013 0.0013 -0.0012 0.0039 0.6696 \n", " 677 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 678 0.0002 0.0001 -0.0000 0.0004 0.0000 ***\n", " 679 0.0001 0.0001 0.0000 0.0002 0.0000 ***\n", " 680 0.0002 0.0002 -0.0002 0.0006 0.0000 ***\n", " 681 0.0000 0.0000 -0.0000 0.0001 0.0000 ***\n", " 682 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 683 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 684 0.0001 0.0000 -0.0000 0.0002 0.0000 ***\n", " 685 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 686 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 687 0.0001 0.0001 -0.0001 0.0002 0.0000 ***\n", " 688 0.0005 0.0005 -0.0005 0.0015 0.0061 ***\n", " 689 0.0010 0.0010 -0.0010 0.0030 0.3922 \n", " 690 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 691 0.0017 0.0017 -0.0016 0.0051 0.9340 \n", " 692 0.0016 0.0016 -0.0016 0.0048 0.8762 \n", " 693 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 694 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 695 0.0001 0.0001 -0.0001 0.0003 0.0000 ***\n", " 696 0.0001 0.0000 -0.0000 0.0002 0.0000 ***\n", " 697 0.0013 0.0013 -0.0012 0.0038 0.6590 \n", " 698 0.0000 0.0001 -0.0001 0.0001 0.0000 ***\n", " 699 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 700 0.0001 0.0001 -0.0001 0.0004 0.0000 ***\n", " 701 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 702 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 703 0.0000 0.0000 -0.0000 0.0001 0.0000 ***\n", " 704 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 705 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 706 0.0003 0.0001 -0.0000 0.0005 0.0000 ***\n", " 707 0.0001 0.0001 -0.0001 0.0004 0.0000 ***\n", " 708 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 709 0.0002 0.0001 -0.0000 0.0005 0.0000 ***\n", " 710 0.0001 0.0001 -0.0001 0.0002 0.0000 ***\n", " 711 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 712 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 713 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 714 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 715 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 716 0.0001 0.0001 -0.0001 0.0002 0.0000 ***\n", " 717 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 718 0.0001 0.0001 -0.0001 0.0002 0.0000 ***\n", " 719 0.0001 0.0001 -0.0001 0.0003 0.0000 ***\n", " 720 0.0001 0.0001 -0.0000 0.0002 0.0000 ***\n", " 721 0.0000 0.0001 -0.0001 0.0001 0.0000 ***\n", " 722 0.0006 0.0005 -0.0004 0.0016 0.0141 **\n", " 723 0.0006 0.0006 -0.0005 0.0017 0.0267 **\n", " 724 0.0001 0.0001 -0.0001 0.0004 0.0000 ***\n", " 725 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 726 0.0001 0.0001 -0.0001 0.0003 0.0000 ***\n", " 727 0.0009 0.0009 -0.0009 0.0027 0.3022 \n", " 728 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 729 0.0000 0.0000 -0.0000 0.0001 0.0000 ***\n", " 730 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 731 0.0000 0.0000 -0.0000 0.0001 0.0000 ***\n", " 732 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 733 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 734 0.0000 0.0000 -0.0000 0.0001 0.0000 ***\n", " 735 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 736 0.0000 0.0000 -0.0000 0.0001 0.0000 ***\n", " 737 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 738 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 739 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 740 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 741 0.0006 0.0006 -0.0006 0.0018 0.0342 **\n", " 742 0.0001 0.0001 -0.0001 0.0003 0.0000 ***\n", " 743 0.0001 0.0001 -0.0001 0.0002 0.0000 ***\n", " 744 0.0010 0.0010 -0.0010 0.0030 0.3878 \n", " 745 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 746 0.0013 0.0013 -0.0012 0.0038 0.6361 \n", " 747 0.0002 0.0002 -0.0002 0.0006 0.0000 ***\n", " 748 0.0001 0.0001 -0.0001 0.0002 0.0000 ***\n", " 749 0.0001 0.0001 0.0000 0.0003 0.0000 ***\n", " 750 0.0001 0.0001 -0.0001 0.0002 0.0000 ***\n", " 751 0.0011 0.0011 -0.0011 0.0034 0.5266 \n", " 752 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 753 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 754 0.0000 0.0001 -0.0001 0.0002 0.0000 ***\n", " 755 0.0002 0.0002 -0.0002 0.0005 0.0000 ***\n", " 756 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 757 0.0001 0.0001 -0.0001 0.0002 0.0000 ***\n", " 758 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 759 0.0014 0.0013 -0.0013 0.0040 0.7000 \n", " 760 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 761 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 762 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 763 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 764 0.0001 0.0000 -0.0000 0.0002 0.0000 ***\n", " 765 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 766 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 767 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 768 0.0003 0.0003 -0.0002 0.0008 0.0000 ***\n", " 769 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 770 0.0001 0.0001 -0.0000 0.0002 0.0000 ***\n", " 771 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 772 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 773 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 774 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 775 0.0001 0.0001 -0.0000 0.0003 0.0000 ***\n", " 776 0.0000 0.0001 -0.0001 0.0002 0.0000 ***\n", " 777 0.0001 0.0001 -0.0000 0.0002 0.0000 ***\n", " 778 0.0007 0.0007 -0.0006 0.0020 0.0740 *\n", " 779 0.0001 0.0000 -0.0000 0.0002 0.0000 ***\n", " 780 0.0001 0.0001 -0.0001 0.0002 0.0000 ***\n", " 781 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 782 0.0001 0.0001 -0.0001 0.0004 0.0000 ***\n", " 783 0.0013 0.0013 -0.0012 0.0039 0.6757 \n", " 784 0.0001 0.0001 -0.0001 0.0002 0.0000 ***\n", " 785 0.0001 0.0001 -0.0001 0.0004 0.0000 ***\n", " 786 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 787 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 788 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 789 0.0001 0.0001 -0.0001 0.0002 0.0000 ***\n", " 790 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 791 0.0001 0.0000 -0.0000 0.0002 0.0000 ***\n", " 792 0.0001 0.0001 -0.0001 0.0003 0.0000 ***\n", " 793 0.0002 0.0001 -0.0001 0.0004 0.0000 ***\n", " 794 0.0003 0.0002 -0.0002 0.0007 0.0000 ***\n", " 795 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 796 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 797 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 798 0.0002 0.0001 -0.0001 0.0004 0.0000 ***\n", " 799 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 800 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 801 0.0001 0.0001 -0.0001 0.0004 0.0000 ***\n", " 802 0.0002 0.0001 0.0001 0.0003 0.0000 ***\n", " 803 0.0001 0.0001 -0.0000 0.0002 0.0000 ***\n", " 804 0.0014 0.0014 -0.0013 0.0040 0.7114 \n", " 805 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 806 0.0029 0.0003 0.0023 0.0036 0.0020 ***\n", " 807 0.0005 0.0001 0.0004 0.0007 0.0000 ***\n", " 808 0.0001 0.0001 -0.0001 0.0003 0.0000 ***\n", " 809 0.0001 0.0001 -0.0001 0.0002 0.0000 ***\n", " 810 0.0000 0.0000 -0.0000 0.0001 0.0000 ***\n", " 811 0.0001 0.0000 -0.0000 0.0002 0.0000 ***\n", " 812 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 813 0.0000 0.0000 -0.0000 0.0001 0.0000 ***\n", " 814 0.0001 0.0001 -0.0001 0.0002 0.0000 ***\n", " 815 0.0015 0.0002 0.0010 0.0019 0.0934 *\n", " 816 0.0010 0.0003 0.0005 0.0016 0.0044 ***\n", " 817 0.0001 0.0001 -0.0001 0.0003 0.0000 ***\n", " 818 0.0001 0.0001 -0.0001 0.0003 0.0000 ***\n", " 819 0.0001 0.0000 -0.0000 0.0002 0.0000 ***\n", " 820 0.0001 0.0000 -0.0000 0.0001 0.0000 ***\n", " 821 0.0004 0.0001 0.0002 0.0006 0.0000 ***\n", " 822 0.0005 0.0005 -0.0004 0.0014 0.0025 ***\n", " 823 0.0002 0.0001 -0.0001 0.0004 0.0000 ***\n", " 824 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 825 0.0002 0.0001 -0.0000 0.0005 0.0000 ***\n", " 826 0.0004 0.0001 0.0002 0.0006 0.0000 ***\n", " 827 0.0003 0.0001 0.0001 0.0005 0.0000 ***\n", " 828 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 829 0.0001 0.0001 -0.0001 0.0004 0.0000 ***\n", " 830 0.0000 0.0000 -0.0001 0.0001 0.0000 ***\n", " 831 0.0001 0.0000 0.0000 0.0002 0.0000 ***\n", "==============================================================================\n", "Significant units: 781 / 832\n", "---\n", "Signif. codes: 0 '***' 0.01 '**' 0.05 '*' 0.1 ' ' 1\n", "==============================================================================\n" ] } ], "source": [ "\n", "print(\"=== Z-space feature summary (primary for FlowExplainer on binary data) ===\")\n", "_ = fdfi_estimator.summary(\n", " alpha=0.05,\n", " target=\"Z\",\n", " var_floor_c=1e-3,\n", " var_floor_method=\"fixed\",\n", " margin=0.0,\n", " margin_method=\"auto\",\n", " margin_quantile=0.95,\n", " alternative=\"two-sided\",\n", ")\n", "\n", "print(\"\\n=== X-space feature summary (secondary — Jacobian-based, may be noisy) ===\")\n", "_ = fdfi_estimator.summary(\n", " alpha=0.05,\n", " target=\"X\",\n", " var_floor_c=1e-3,\n", " var_floor_method=\"fixed\",\n", " margin=0.0,\n", " margin_method=\"auto\",\n", " margin_quantile=0.95,\n", " alternative=\"two-sided\",\n", ")\n" ] }, { "cell_type": "markdown", "id": "233d4096", "metadata": {}, "source": [ "## Group Importance\n" ] }, { "cell_type": "code", "execution_count": 8, "id": "52091cce", "metadata": {}, "outputs": [], "source": [ "res_df = fdfi_estimator.conf_int(\n", " alpha=0.05,\n", " target=\"Z\",\n", " groups=df_groups,\n", " threshold_null=True,\n", " var_floor_c=1e-3,\n", " var_floor_method=\"fixed\",\n", " margin=0.0,\n", " margin_method=\"fixed\",\n", " alternative=\"two-sided\",\n", " multitest_method=\"bonferroni\",\n", " verbose=False,\n", ")\n", "res_df = pd.DataFrame({\n", " \"group\": res_df[\"groups\"],\n", " \"importance\": res_df[\"score\"],\n", " \"se\": res_df[\"se\"],\n", " \"zscore\": res_df[\"zscore\"],\n", " \"ci_lower\": res_df[\"ci_lower\"],\n", " \"ci_upper\": res_df[\"ci_upper\"],\n", " \"p_value\": res_df[\"pvalue\"],\n", " \"p_value_adj\": res_df.get(\"pvalue_adj\", res_df[\"pvalue\"]),\n", " \"reject_null\": res_df[\"reject_null\"],\n", "})\n" ] }, { "cell_type": "markdown", "id": "5e29a189", "metadata": {}, "source": [ "## Group Summary Table\n" ] }, { "cell_type": "code", "execution_count": 9, "id": "8dffe80d", "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "=== Z-space group summary (primary for FlowExplainer) ===\n", "==============================================================================\n", "Feature Importance Results\n", "==============================================================================\n", "Method: FlowExplainer\n", "Number of units: 14\n", "Significance level: 0.05\n", "Alternative: two-sided\n", "Multiple testing: bonferroni\n", "Margin method: fixed\n", "------------------------------------------------------------------------------\n", " Group Estimate Std Err CI Lower CI Upper Adj P-val Sig\n", "------------------------------------------------------------------------------\n", " cd4bs 0.0434 0.0330 -0.0213 0.1081 1.0000 \n", " covar 0.0197 0.0125 -0.0049 0.0443 1.0000 \n", " cysteines 0.0012 0.0006 -0.0001 0.0024 0.9081 \n", " esa 0.0288 0.0209 -0.0122 0.0697 1.0000 \n", " geog 0.0003 0.0002 -0.0001 0.0006 1.0000 \n", " geometry 0.0028 0.0005 0.0019 0.0038 0.0000 ***\n", " glyco 0.0049 0.0020 0.0010 0.0088 0.1884 \n", " gp41 0.0101 0.0087 -0.0069 0.0270 1.0000 \n", " pngs 0.0184 0.0078 0.0031 0.0337 0.2563 \n", " pngs_novrc01 0.0134 0.0116 -0.0093 0.0360 1.0000 \n", " sequons 0.0037 0.0004 0.0029 0.0045 0.0000 ***\n", " steric_bulk 0.0009 0.0002 0.0004 0.0013 0.0017 ***\n", " subtype 0.0023 0.0018 -0.0013 0.0059 1.0000 \n", " vrc01 0.0396 0.0280 -0.0154 0.0945 1.0000 \n", "==============================================================================\n", "Significant units: 3 / 14\n", "---\n", "Signif. codes: 0 '***' 0.01 '**' 0.05 '*' 0.1 ' ' 1\n", "==============================================================================\n", "\n", "=== X-space group summary (secondary — Jacobian-based) ===\n", "==============================================================================\n", "Feature Importance Results\n", "==============================================================================\n", "Method: FlowExplainer\n", "Number of units: 14\n", "Significance level: 0.05\n", "Alternative: two-sided\n", "Multiple testing: bonferroni\n", "Margin method: fixed\n", "------------------------------------------------------------------------------\n", " Group Estimate Std Err CI Lower CI Upper Adj P-val Sig\n", "------------------------------------------------------------------------------\n", " cd4bs 0.0372 0.0276 -0.0168 0.0912 1.0000 \n", " covar 0.0200 0.0124 -0.0042 0.0442 1.0000 \n", " cysteines 0.0011 0.0006 -0.0001 0.0022 0.8767 \n", " esa 0.0254 0.0181 -0.0101 0.0609 1.0000 \n", " geog 0.0003 0.0002 -0.0000 0.0006 0.8746 \n", " geometry 0.0028 0.0005 0.0018 0.0038 0.0000 ***\n", " glyco 0.0052 0.0022 0.0009 0.0096 0.2560 \n", " gp41 0.0093 0.0079 -0.0062 0.0249 1.0000 \n", " pngs 0.0193 0.0087 0.0023 0.0364 0.3686 \n", " pngs_novrc01 0.0121 0.0101 -0.0077 0.0319 1.0000 \n", " sequons 0.0039 0.0004 0.0031 0.0048 0.0000 ***\n", " steric_bulk 0.0009 0.0002 0.0004 0.0014 0.0035 ***\n", " subtype 0.0020 0.0016 -0.0011 0.0051 1.0000 \n", " vrc01 0.0344 0.0238 -0.0123 0.0810 1.0000 \n", "==============================================================================\n", "Significant units: 3 / 14\n", "---\n", "Signif. codes: 0 '***' 0.01 '**' 0.05 '*' 0.1 ' ' 1\n", "==============================================================================\n" ] } ], "source": [ "\n", "print(\"=== Z-space group summary (primary for FlowExplainer) ===\")\n", "_ = fdfi_estimator.summary(\n", " alpha=0.05,\n", " target=\"Z\",\n", " groups=df_groups,\n", " threshold_null=True,\n", " var_floor_c=1e-3,\n", " var_floor_method=\"fixed\",\n", " margin=0.0,\n", " margin_method=\"fixed\",\n", " alternative=\"two-sided\",\n", " multitest_method=\"bonferroni\",\n", ")\n", "print(\"\\n=== X-space group summary (secondary — Jacobian-based) ===\")\n", "_ = fdfi_estimator.summary(\n", " alpha=0.05,\n", " target=\"X\",\n", " groups=df_groups,\n", " threshold_null=True,\n", " var_floor_c=1e-3,\n", " var_floor_method=\"fixed\",\n", " margin=0.0,\n", " margin_method=\"fixed\",\n", " alternative=\"two-sided\",\n", " multitest_method=\"bonferroni\",\n", ")\n" ] }, { "cell_type": "markdown", "id": "4f835028", "metadata": {}, "source": [ "## Group Mapping\n" ] }, { "cell_type": "code", "execution_count": 10, "id": "4d12c4f4", "metadata": {}, "outputs": [], "source": [ "group_mapping = pd.DataFrame({\n", " \"group_num\": range(1, 15),\n", " \"group\": [\n", " \"vrc01\", \"cd4bs\", \"esa\", \"glyco\", \"covar\", \"pngs\", \"gp41\",\n", " \"pngs_novrc01\", \"subtype\", \"sequons\", \"geometry\", \"cysteines\",\n", " \"steric_bulk\", \"geog\"\n", " ],\n", " \"description\": [\n", " \"VRC01 binding footprint\",\n", " \"CD4 binding sites\",\n", " \"Sites with sufficient exposed surface area\",\n", " \"Sites identified as important for glycosylation\",\n", " \"Sites with residues that covary with\\nthe VRC01 binding footprint\",\n", " \"Sites associated with VRC01-specific\\npotential N-linked glycosylation (PNGS) effects\",\n", " \"gp41 sites important for VRC01 binding\",\n", " \"Sites for indicating N-linked glycosylation\",\n", " \"Majority virus subtypes\",\n", " \"Region-specific counts of PNGS\",\n", " \"Viral geometry\",\n", " \"Cysteine counts\",\n", " \"Steric bulk at critical locations\",\n", " \"Geographic confounders\",\n", " ],\n", "})\n" ] }, { "cell_type": "markdown", "id": "bbaef204", "metadata": {}, "source": [ "## Group Importance Plot\n" ] }, { "cell_type": "code", "execution_count": 11, "id": "0a943bfe", "metadata": {}, "outputs": [ { "data": { "image/png": "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", "text/plain": [ "
" ] }, "metadata": {}, "output_type": "display_data" } ], "source": [ "\n", "DATASET_LABELS = {\n", " \"sens50\": \"Sensitivity 50\",\n", " \"sens80\": \"Sensitivity 80\",\n", " \"ic50\": \"IC50 Censored\",\n", "}\n", "dataset_label = DATASET_LABELS.get(outcome, outcome)\n", "\n", "group_df_sorted = res_df.sort_values(\"importance\", ascending=True).reset_index(drop=True)\n", "group_df_sorted = group_df_sorted.merge(group_mapping, on=\"group\", how=\"left\")\n", "\n", "fig, axes = plt.subplots(1, 3, figsize=(18, 6), gridspec_kw={\"width_ratios\": [3, 3, 3]})\n", "fig.suptitle(\n", " f\"{dataset_label} ({outcome}) - Flow FDFI Group Importance (Z-space)\",\n", " fontsize=22,\n", " y=1.04,\n", ")\n", "\n", "group_numbers = list(group_df_sorted[\"group_num\"])\n", "\n", "ax1 = axes[0]\n", "ax1.barh(\n", " range(len(group_df_sorted)),\n", " group_df_sorted[\"importance\"],\n", " xerr=group_df_sorted[\"se\"],\n", " error_kw=dict(capsize=5),\n", " alpha=0.7,\n", " color=\"steelblue\",\n", ")\n", "ax1.set_yticks(range(len(group_df_sorted)))\n", "ax1.set_yticklabels(group_numbers, fontsize=12)\n", "ax1.set_xlabel(\"Group Importance\", fontsize=16)\n", "ax1.set_title(\"Feature Group Importance (Flow FDFI, Z)\", fontsize=18)\n", "ax1.set_ylabel(\"Group Number\", fontsize=16)\n", "ax1.grid(axis=\"x\", alpha=0.3)\n", "\n", "for i, (_, row) in enumerate(group_df_sorted.iterrows()):\n", " if row[\"reject_null\"]:\n", " ax1.text(\n", " row[\"importance\"] + row[\"se\"] + 0.001,\n", " i,\n", " \"*\",\n", " va=\"center\",\n", " fontweight=\"bold\",\n", " color=\"red\",\n", " )\n", "\n", "ax2 = axes[1]\n", "colors = [\"red\" if reject_null else \"gray\" for reject_null in group_df_sorted[\"reject_null\"]]\n", "ax2.barh(range(len(group_df_sorted)), group_df_sorted[\"zscore\"], alpha=0.7, color=colors)\n", "ax2.set_yticks(range(len(group_df_sorted)))\n", "ax2.set_yticklabels(group_numbers, fontsize=12)\n", "ax2.set_xlabel(\"Z-score\", fontsize=16)\n", "ax2.set_title(\"Feature Group Z-scores\", fontsize=18)\n", "ax2.grid(axis=\"x\", alpha=0.3)\n", "ax2.axvline(x=1.96, color=\"red\", linestyle=\"--\", linewidth=2, alpha=0.7, label=\"p=0.05\")\n", "ax2.legend(fontsize=12)\n", "\n", "ax3 = axes[2]\n", "ax3.set_xlim(0, 1)\n", "ax3.set_ylim(-0.5, len(group_df_sorted) - 0.5)\n", "\n", "for i, (_, row) in enumerate(group_df_sorted.iterrows()):\n", " group_name = row[\"group\"]\n", " display_name = row[\"description\"] if pd.notna(row[\"description\"]) else group_name\n", " text_color = \"red\" if row[\"reject_null\"] else \"gray\"\n", "\n", " ax3.text(\n", " 0.02,\n", " i,\n", " f\"{int(row['group_num']):2d}.\",\n", " va=\"center\",\n", " ha=\"left\",\n", " fontsize=14,\n", " fontweight=\"bold\",\n", " color=\"black\",\n", " )\n", " ax3.text(\n", " 0.09,\n", " i,\n", " display_name,\n", " va=\"center\",\n", " ha=\"left\",\n", " fontsize=12,\n", " color=text_color,\n", " )\n", "\n", "ax3.set_yticks([])\n", "ax3.set_xticks([])\n", "ax3.set_title(\"Feature Group Annotations\", fontsize=18)\n", "for spine in ax3.spines.values():\n", " spine.set_visible(False)\n", "\n", "plt.tight_layout(rect=[0, 0, 1, 0.95])\n", "plt.show()\n" ] }, { "cell_type": "markdown", "id": "fa0fb186", "metadata": {}, "source": [ "## Visualize and Compare Results: Flow vs EOT\n" ] }, { "cell_type": "code", "execution_count": 12, "id": "75a2ba92", "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "=== Group-level results (Flow FDFI, Z-space) ===\n", "Group Importance SE Z-score Significant\n", "----------------------------------------------------------------------\n", "sequons 0.0037 0.0004 9.20 ***\n", "geometry 0.0028 0.0005 5.78 ***\n", "steric_bulk 0.0009 0.0002 3.84 ***\n", "glyco 0.0049 0.0020 2.47 \n", "pngs 0.0184 0.0078 2.36 \n", "cysteines 0.0012 0.0006 1.85 \n", "geog 0.0003 0.0002 1.62 \n", "covar 0.0197 0.0125 1.57 \n", "vrc01 0.0396 0.0280 1.41 \n", "esa 0.0288 0.0209 1.38 \n", "cd4bs 0.0434 0.0330 1.31 \n", "subtype 0.0023 0.0018 1.23 \n", "gp41 0.0101 0.0087 1.16 \n", "pngs_novrc01 0.0134 0.0116 1.16 \n", "\n", "Significant groups: 3/14 (Bonferroni alpha=0.05)\n", "Z-score range: 1.16 to 9.20\n" ] } ], "source": [ "print(\"=== Group-level results (Flow FDFI, Z-space) ===\")\n", "print(f\"{'Group':<20} {'Importance':>12} {'SE':>10} {'Z-score':>10} {'Significant':>12}\")\n", "print(\"-\" * 70)\n", "for _, row in res_df.sort_values(\"zscore\", ascending=False).iterrows():\n", " sig = \"***\" if row[\"reject_null\"] else \"\"\n", " print(f\"{row['group']:<20} {row['importance']:>12.4f} {row['se']:>10.4f} {row['zscore']:>10.2f} {sig:>12}\")\n", "\n", "n_sig = res_df[\"reject_null\"].sum()\n", "print(f\"\\nSignificant groups: {n_sig}/{len(res_df)} (Bonferroni alpha=0.05)\")\n", "print(f\"Z-score range: {res_df['zscore'].min():.2f} to {res_df['zscore'].max():.2f}\")\n" ] }, { "cell_type": "markdown", "id": "7590aedc", "metadata": {}, "source": [ "## Note on Variance Floor (`var_floor_c`)\n", "\n", "The default `var_floor_c=0.1` used by EOTExplainer is calibrated for the OT/EOT\n", "linear decoder, where the UEIF variance across samples is naturally large.\n", "FlowExplainer uses a CPI-based UEIF, $\\text{UEIF}_{i,j} = (y_i - \\bar{\\tilde{y}}_{ij})^2$,\n", "whose value is nearly constant across samples $i$ (the nonlinear flow decoder maps\n", "similar latent perturbations to the same binary output pattern for most samples).\n", "This makes the raw per-sample standard deviation structurally small — not due to\n", "noise, but by construction.\n", "\n", "Using `var_floor_c=1e-3` sets a floor of $10^{-3}/\\sqrt{n} \\approx 4 \\times 10^{-5}$,\n", "which is roughly 3× below the smallest observed group SE and adds less than 5% to\n", "any z-score. It guards against numerical zeros at the individual-feature level\n", "while leaving all group-level inferences essentially unchanged." ] } ], "metadata": { "kernelspec": { "display_name": "Python 3", "language": "python", "name": "python3" }, "language_info": { "codemirror_mode": { "name": "ipython", "version": 3 }, "file_extension": ".py", "mimetype": "text/x-python", "name": "python", "nbconvert_exporter": "python", "pygments_lexer": "ipython3", "version": "3.10.19" } }, "nbformat": 4, "nbformat_minor": 5 }