--- a
+++ b/notebooks/stats.ipynb
@@ -0,0 +1,159 @@
+{
+ "cells": [
+  {
+   "cell_type": "code",
+   "execution_count": 27,
+   "metadata": {
+    "scrolled": false
+   },
+   "outputs": [
+    {
+     "data": {
+      "text/plain": [
+       "                  dataset        cutoff    algorithm    rank_auc   \n",
+       " geis_250.00          :30   1 visit :210   LR :210   Min.   : 1.0  \n",
+       " geis_250.00_cutoff182:30   1 year  :210   RF :210   1st Qu.: 8.0  \n",
+       " geis_250.00_cutoff365:30   6 months:210   XGB:210   Median :15.5  \n",
+       " geis_250.40          :30                            Mean   :15.5  \n",
+       " geis_250.40_cutoff182:30                            3rd Qu.:23.0  \n",
+       " geis_250.40_cutoff365:30                            Max.   :30.0  \n",
+       " geis_327.23          :30                                          \n",
+       " geis_327.23_cutoff182:30                                          \n",
+       " geis_327.23_cutoff365:30                                          \n",
+       " geis_331.0           :30                                          \n",
+       " geis_331.0_cutoff182 :30                                          \n",
+       " geis_331.0_cutoff365 :30                                          \n",
+       " geis_530.81          :30                                          \n",
+       " geis_530.81_cutoff182:30                                          \n",
+       " geis_530.81_cutoff365:30                                          \n",
+       " geis_571.8           :30                                          \n",
+       " geis_571.8_cutoff182 :30                                          \n",
+       " geis_571.8_cutoff365 :30                                          \n",
+       " geis_585.9           :30                                          \n",
+       " geis_585.9_cutoff182 :30                                          \n",
+       " geis_585.9_cutoff365 :30                                          "
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    }
+   ],
+   "source": [
+    "setwd('/media/bill/Drive/projects/geis-ehr/analysis')\n",
+    "\n",
+    "df <- read.csv(\"auc_rankings.csv\",header=TRUE,sep=',')\n",
+    "# df <- subset(df,!is.na(rank))\n",
+    "# df <- subset(df,dataset!='505_tecator')\n",
+    "# df <- subset(df,algorithm != 'LR')\n",
+    "summary(df,maxsum=21)\n"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 29,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/plain": [
+       "\n",
+       "\tPairwise comparisons using Wilcoxon signed rank test \n",
+       "\n",
+       "data:  dfs$rank_auc and dfs$algorithm \n",
+       "\n",
+       "    LR      RF     \n",
+       "RF  1.0e-12 -      \n",
+       "XGB 1.1e-12 7.8e-12\n",
+       "\n",
+       "P value adjustment method: bonferroni "
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    }
+   ],
+   "source": [
+    "dfs <- subset(df,cutoff=='1 visit')\n",
+    "pairwise.wilcox.test(dfs$rank_auc, dfs$algorithm, p.adjust.method = 'bonferroni',\n",
+    "                      paired = T)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 31,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/plain": [
+       "\n",
+       "\tPairwise comparisons using Wilcoxon signed rank test \n",
+       "\n",
+       "data:  dfs$rank_auc and dfs$algorithm \n",
+       "\n",
+       "    LR      RF     \n",
+       "RF  1.0e-12 -      \n",
+       "XGB 1.0e-12 2.3e-12\n",
+       "\n",
+       "P value adjustment method: bonferroni "
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    }
+   ],
+   "source": [
+    "dfs <- subset(df,cutoff=='6 months')\n",
+    "pairwise.wilcox.test(dfs$rank_auc, dfs$algorithm, p.adjust.method = 'bonferroni',\n",
+    "                      paired = T)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 28,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/plain": [
+       "\n",
+       "\tPairwise comparisons using Wilcoxon signed rank test \n",
+       "\n",
+       "data:  dfs$rank_auc and dfs$algorithm \n",
+       "\n",
+       "    LR      RF     \n",
+       "RF  1.0e-12 -      \n",
+       "XGB 1.0e-12 9.5e-12\n",
+       "\n",
+       "P value adjustment method: bonferroni "
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    }
+   ],
+   "source": [
+    "dfs <- subset(df,cutoff=='1 year')\n",
+    "pairwise.wilcox.test(dfs$rank_auc, dfs$algorithm, p.adjust.method = 'bonferroni',\n",
+    "                      paired = T)"
+   ]
+  }
+ ],
+ "metadata": {
+  "kernelspec": {
+   "display_name": "R",
+   "language": "R",
+   "name": "ir"
+  },
+  "language_info": {
+   "codemirror_mode": "r",
+   "file_extension": ".r",
+   "mimetype": "text/x-r-source",
+   "name": "R",
+   "pygments_lexer": "r",
+   "version": "3.5.1"
+  }
+ },
+ "nbformat": 4,
+ "nbformat_minor": 2
+}