cooksd.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "import numpy as np\n",
    "from numpy.linalg import inv\n",
    "import matplotlib\n",
    "import matplotlib.pyplot as plt\n",
    "from sklearn.metrics import mean_squared_error"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "def dataset_ (N=200, idx_outlier=0, ydistance=5):\n",
    "    rng = np.random.RandomState(4)\n",
    "    data  = np.dot(rng.rand(2, 2), rng.randn(2, N)).T\n",
    "    data[idx_outlier:idx_outlier+1,1] = ydistance\n",
    "    return data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "N=50\n",
    "inx=10\n",
    "i = dataset_(N, inx)\n",
    "Z = i[:,0].reshape(len(i[:,0]),1)\n",
    "y = i[:,1].reshape(len(i[:,1]),1)\n",
    "b = lambda Z,y: inv(np.dot(Z.T,Z)).dot(Z.T).dot(y)\n",
    "ypred = lambda Z,b: Z.dot(b)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "original_ypred = ypred(Z, b(Z,y))\n",
    "p = len(i.T)-1\n",
    "s2 = mean_squared_error(y, original_ypred)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "outliers,Ds = [],[]\n",
    "t=2*(p+1)/N\n",
    "for n in range(N):\n",
    "    #without i-th obeservation\n",
    "    tempZ = np.delete(Z, (n), axis=0)\n",
    "    tempY = np.delete(y, (n), axis=0)\n",
    "    tempB = b(tempZ, tempY)\n",
    "    tempYPred = ypred(tempZ, tempB)\n",
    "    #get rid of i-th observation from oritinal predition too\n",
    "    original_YPred_ = np.delete(original_ypred, (n), axis=0)\n",
    "    \n",
    "    errs = sum((original_YPred_ - tempYPred)**2)\n",
    "    D = errs/(p*s2)\n",
    "    Ds.append(D)\n",
    "    if D[0] > t: \n",
    "        outliers.append(n)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[array([0.1488515]),\n",
       " array([0.00753585]),\n",
       " array([0.00690743]),\n",
       " array([0.00671447]),\n",
       " array([0.0046191])]"
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "Ds.sort(reverse=True);Ds[:5]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[10]"
      ]
     },
     "execution_count": 7,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "outliers"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "hide_input": false,
  "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": "3.5.6"
  },
  "toc": {
   "base_numbering": 1,
   "nav_menu": {},
   "number_sections": true,
   "sideBar": true,
   "skip_h1_title": false,
   "title_cell": "Table of Contents",
   "title_sidebar": "Contents",
   "toc_cell": false,
   "toc_position": {},
   "toc_section_display": true,
   "toc_window_display": false
  },
  "varInspector": {
   "cols": {
    "lenName": 16,
    "lenType": 16,
    "lenVar": 40
   },
   "kernels_config": {
    "python": {
     "delete_cmd_postfix": "",
     "delete_cmd_prefix": "del ",
     "library": "var_list.py",
     "varRefreshCmd": "print(var_dic_list())"
    },
    "r": {
     "delete_cmd_postfix": ") ",
     "delete_cmd_prefix": "rm(",
     "library": "var_list.r",
     "varRefreshCmd": "cat(var_dic_list()) "
    }
   },
   "types_to_exclude": [
    "module",
    "function",
    "builtin_function_or_method",
    "instance",
    "_Feature"
   ],
   "window_display": false
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}