{ "cells": [ { "cell_type": "markdown", "metadata": {}, "source": [ "# Lab 1.5\n", "In this homework, the exercises are based on the problems solved in the pre-asssement test." ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "## Set-up" ] }, { "cell_type": "code", "execution_count": 1, "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "Modules Imported!\n" ] } ], "source": [ "%matplotlib inline\n", "import matplotlib.pyplot as plt\n", "import numpy as np\n", "import scipy as sp\n", "import scipy.stats as st\n", "print(\"Modules Imported!\")" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "## Ploting pdf and CDF" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "In Python, distributions and related functions can be created using scipy.stat. The general form is (using st for scipy.stat, assuming 2 parameters):\n", "python\n", "myRV = st.RVName(param1,param2,loc=0,scale=1)\n", "\n", "Enter the loc and scale parameters if they are different from the default. You can then get the pdf at x as\n", "python\n", "myRV.pdf(x)\n", "\n", "We'll see this for the Beta distribution below. The Beta distributon has two parameters $\\alpha$ and $\\beta$ (called shape parameters)." ] }, { "cell_type": "code", "execution_count": 2, "metadata": {}, "outputs": [ { "data": { "text/plain": [ "[]" ] }, "execution_count": 2, "metadata": {}, "output_type": "execute_result" }, { "data": { "image/png": "iVBORw0KGgoAAAANSUhEUgAAAXcAAAD8CAYAAACMwORRAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nCyncRUS60P8HP0CMxdvk1OQAAAAASUVORK5CYII=\n", "text/plain": [ "" ] }, "metadata": {}, "output_type": "display_data" }, { "data": { "image/png": "iVBORw0KGgoAAAANSUhEUgAAAXcAAAD8CAYAAACMwORRAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz\nAAALEgAACxIB0t1+/AAAGY9JREFUeJzt3XuMXOd53/Hvs/cb98LdFbnkLrmKRQWmEtmSGMax2liA\nm1oS2ioXu7UQxIiDQHFgIzaSog0SwEadpkDaIg1sJ1YFWE0NuE4d2AmEhI4To259aWiLohlZd1OU\nKC5Jk6vdJfc+szPz5I+ZWY5Xu5zbe+acmf19gIV2Zw53ngHn/PTyOe97XnN3RESktbTFXYCIiISn\ncBcRaUEKdxGRFqRwFxFpQQp3EZEWpHAXEWlBCncRkRakcBcRaUEKdxGRFtQR1wuPjY359PR0XC8v\nItKUnnrqqdfdfbzccbGF+/T0NKdOnYrr5UVEmpKZna/kOLVlRERakMJdRKQFKdxFRFqQwl1EpAUp\n3EVEWpDCXUSkBSncRURakMJdYnFlcZ1P/9+XeenKUtyliLQkhbs03Go6w8//yf/nD/7mBd7z6N9z\nYX417pJEWo7CXRrus39/novX1vjP776TVCbLf/vqS3GXJNJyFO7ScF86PcPx6b3862NTPHz8EE+c\nucTCSjruskRaisJdGurc7DIvXVnmwR/fD8Av3D1JJueceOZyzJWJtBaFuzTUyXPzALzjR28B4I4D\ngxza28fXXrgaZ1kiLUfhLg116vw8o/1dTI/2AWBm3HvbGN8+N08mm4u5OpHWoXCXhjrz2jXuPjyC\nmW0+du9toyylMjx98XqMlYm0lrLhbmZTZvY1M3vOzJ41sw9vc4yZ2SfM7KyZPW1md0dTrjSz9Y0s\nr86t8Ob9e37o8Z/6kVEATp6bi6MskZZUyWYdGeC33P20me0BnjKzv3P350qOeQA4Uvj6SeDThf+K\nbHp5dpmcw+1bwn10oJvp0T7+4cK1mCoTaT1lR+7uftndTxe+XwKeBw5uOewh4LOedxIYNrOJ4NVK\nU/v+lWUAbt+35w3P/fjkMN+bUVtGJJSqeu5mNg3cBXx7y1MHgQslP8/wxv8BYGaPmNkpMzs1Oztb\nXaXS9F68skRnuzE92v+G594yOcSl6+vMLqViqEyk9VQc7mY2AHwR+Ii7L9byYu7+mLsfc/dj4+Nl\n93eVFvP9K0vcOtZPV8cbP3Z3Tg4D8PSMWjMiIVQU7mbWST7YP+fuX9rmkIvAVMnPk4XHRDa9Nr+6\n7agd4OiBQQCev1zTuEFEtqhktowBnwGed/c/3OGwJ4D3FWbNvA247u5aciib3J2ZhTUmR/q2fX6g\nu4ODw728VOjLi0h9Kpktcy/wS8D3zOxM4bHfAQ4BuPujwAngQeAssAq8P3yp0swWVjdYTWeZHOnd\n8Zjb9w3oFsAigZQNd3f/JmBljnHgg6GKktYzs5C/re/Nw30P3zo7Ryabo6Nd6+tE6qEzSBpiZmEN\nYMe2DOTDPZ3N8eqc7u8uUi+FuzREceR+sMzIHVBrRiQAhbs0xMzCGoM9HQz1du54zG23DGCmcBcJ\nQeEuDXHp2hoHhncetQP0drUzOdLL2auaMSNSL4W7NMTVpRTje7rLHjc92s959dxF6qZwl4aYXUpx\ny56essfdOtbPq3Mr5CdgiUitFO4SuVzO8+E+WH7kfni0n6X1DPPaU1WkLgp3idzCappMzrmlgrbM\nrWP5qZKaDilSH4W7RO5q4U6PlbRliveeefX1lUhrEml1CneJ3Ga4V9CWmRzpo83g/JzCXaQeCneJ\n3NXFdYCK2jJdHW1MjvTxitoyInVRuEvkiiP3SqZCAhwe7dPIXaROCneJ3OxSiv6udvq6KrkJaX46\n5CuvazqkSD0U7hK5hdU0owOVjdohf1FV0yFF6qNwl8jNr6TZ299V8fGH9uanQxbvJCki1VO4S+QW\nVqsL98m9+XvQKNxFaqdwl8jNL6cZ6as83A8OF8NdM2ZEaqVwl8jNr6bZ27/zrX632tPTyVBvp0bu\nInVQuEuk1tJZ1jdyjFTRloH8dnwauYvUTuEukZpfzc942VtFWwaK4a6Ru0itFO4SqYXCdMZqLqhC\n/jYEMwtrmusuUiOFu0RqvuZw72VtI6u57iI1UrhLpBYKbZlqe+43ZsyoNSNSC4W7RGpz5F51z10L\nmUTqoXCXSC2sbgAw2Fv5VEiAgyOa6y5SD4W7RGpxbYM9PR20t1lVf26ot5PBng4uXtPIXaQWCneJ\n1OLaBkNVjtqLijNmRKR6CneJ1OL6BoM9tYa7FjKJ1ErhLpFaXMsw2FvZfdy3OlhYyKS57iLVU7hL\npOoZuR8Y6mU1nWVxLRO4KpHWp3CXSNXTc58Y7gHg0nX13UWqpXCXSF1f26h6GmTRxFB+OuRlhbtI\n1RTuEplMNsdKOlt7W6Y4cr+2HrIskV1B4S6RWVrP98prvaB6y54e2ttMI3eRGijcJTKL64XVqTWO\n3NvbjH17urmskbtI1RTuEpnra/lwr/WCKsDEcK8uqIrUQOEukSlOYaz1girAxFAPl69r5C5SLYW7\nRGazLVNjzx3gwHAvl6+vayGTSJXKhruZPW5mV83smR2ev8/MrpvZmcLXR8OXKc1oca2+njvkR+7p\nTI45bdohUpVKRu5/Ctxf5phvuPtbC18fr78saQXFkXtdPffiXHddVBWpStlwd/evA/MNqEVazPW1\nDdrbjL6u9pp/xwGtUhWpSaie+9vN7Gkz+7KZ3bHTQWb2iJmdMrNTs7OzgV5akmpxLcNgTwdm1d3L\nvdSNkbvCXaQaIcL9NHDI3e8EPgn85U4Huvtj7n7M3Y+Nj48HeGlJssX12m89UDTa30VXe5tmzIhU\nqe5wd/dFd18ufH8C6DSzsbork6a3uFb7HSGL2tqM/UM9XFK4i1Sl7nA3s/1W+He3mR0v/M65en+v\nNL+VVJb+7tr77UUTQz1qy4hUqewEZDP7PHAfMGZmM8DHgE4Ad38UeDfw62aWAdaA97omJQuwnMps\nXhCtx4HhXr7ziq7pi1SjbLi7+8Nlnv8U8KlgFUnLWEln6O+ufQFT0cRQD1cW18nmvOqNtkV2K61Q\nlcispAKF+3AvmZzz+nIqQFUiu4PCXSKznMowECDcDwwV7+uuvrtIpRTuEolMNsf6Ri5IuO8vhLum\nQ4pUTuEukVhJZQGCtGUOFBYyaeQuUjmFu0RiOZ2/3e9AgKmQw32d9HS28QON3EUqpnCXSKyk8uEe\nYuRuZkwM9aotI1IFhbtEYjlguENx0w61ZUQqpXCXSBRH7iEuqEL+oqraMiKVU7hLJDbbMl1hwv3A\nUC9XllJkc1r8LFIJhbtEYrkwWybkyD2bc2aXtJBJpBIKd4nEjQuq9c+WAW3aIVIthbtEIvQF1f2D\n+bnu6ruLVEbhLpFYSWXoaDO6O8J8xCZ0CwKRqijcJRLFm4bVs8VeKS1kEqmOwl0isZzKBruYClrI\nJFIthbtEYiXQHSFLaSGTSOUU7hKJ/EYdYWbKFGkhk0jlFO4SiaX1MBt1lNJCJpHKKdwlElG0ZbSQ\nSaRyCneJRKgt9kptTodU312kLIW7RGIlnaW/K2zPfWJIC5lEKqVwl0ispbP0BrppWJEWMolUTuEu\nwWWyOdLZHH2BR+5ayCRSOYW7BLe2kb8jZOhw31zItKhwFylH4S7BraXz4d4bONyhsJBJbRmRshTu\nEtxqMdw7w4e7FjKJVEbhLsEVwz10Wwa0kEmkUgp3CW5tI38v99CzZUALmUQqpXCX4NbSOSCakbsW\nMolURuEuwa2mCyP3CHruWsgkUhmFuwRXnAoZ1WwZ0EImkXIU7hJclBdUtZBJpDIKdwmuOM+9rzP8\nBVUtZBKpjMJdgouyLQNayCRSCYW7BLeaztDRZnR1RPPx0kImkfIU7hLcajobyUyZoomhHi1kEilD\n4S7B5W/3G2W492ohk0gZCncJbm0jG8lMmSItZBIpr2y4m9njZnbVzJ7Z4Xkzs0+Y2Vkze9rM7g5f\npjST1Qg26iilhUwi5VUycv9T4P6bPP8AcKTw9Qjw6frLkma2lm7QyF0zZkR2VDbc3f3rwPxNDnkI\n+KznnQSGzWwiVIHSfFbTmUgvqGohk0h5IXruB4ELJT/PFB57AzN7xMxOmdmp2dnZAC8tSbQa8QVV\nLWQSKa+hF1Td/TF3P+bux8bHxxv50tJA6xFfUAXYP6iFTCI3EyLcLwJTJT9PFh6TXWo14p47wMSw\nFjKJ3EyIcH8CeF9h1szbgOvufjnA75UmtZbO0hvBfWVKaSGTyM2VPQPN7PPAfcCYmc0AHwM6Adz9\nUeAE8CBwFlgF3h9VsZJ87s7qRpbermg7fqULmfYXZs+IyA1lw93dHy7zvAMfDFaRNLV0Nkc25/RF\nOM8dfnghk8Jd5I20QlWCWi9ssRflVEjQQiaRchTuEtRqYXPsyC+oFkbrlxXuIttSuEtQxY06opzn\nDjcWMmk6pMj2FO4S1PpGvi3T3RFtuGshk8jNKdwlqOIuTD2d0X+0tJBJZGcKdwkqtRnu0Y7cQQuZ\nRG5G4S5BrWcaGO5ayCSyI4W7BFXsuTeiLaMdmUR2pnCXoNaLbZmIL6iCdmQSuRmFuwR1Y+TeiHDX\nQiaRnSjcJaj1Bs6W0UImkZ0p3CWoRl5QHe7rpLtDC5lEtqNwl6DWN3KYQXdH9B8tM+PAsBYyiWxH\n4S5BpTaydHe0YWYNeT0tZBLZnsJdglrbyDakJVM0OdLLRYW7yBso3CWo9Y1sQ6ZBFk2O9HFlMUWq\n0OsXkTyFuwS1vpFryEyZosmR/HTIS9fUdxcppXCXoNZjaMsAXJhfbdhrijQDhbsEtZ7J0d3IcN/b\nB8DMgvruIqUU7hJUvufeuI/Vvj3ddLQZMwsauYuUUrhLUKkGt2U62tuYGO7RyF1kC4W7BNXoC6oA\nUyN9GrmLbKFwl6DWM40duUP+oqpG7iI/TOEuQTV6njvk57pfXUpt3rRMRBTuEthaOtvwtsyNue4a\nvYsUKdwlqPVMjp6uxo/cQdMhRUop3CWYXM5JZ3IxtGXyI3eFu8gNCncJJpVp3C5MpfYN9tDZrrnu\nIqUU7hJMI3dhKtXelr+vu0buIjco3CWYRu7CtFV+OqRG7iJFCncJ5sbm2I3/WE0O93FBI3eRTQp3\nCWazLdPgC6oAh0b7mF1KsZrONPy1RZJI4S7B3Oi5Nz7cp0f7ATg/p9aMCCjcJaBiW6Y7hrbM4dH8\nXPfzcysNf22RJFK4SzBxjtyL4f6qRu4igMJdAoqz576np5PR/i6N3EUKFO4SzI1FTPF8rA6P9vHq\n6xq5i0CF4W5m95vZi2Z21sx+e5vn7zOz62Z2pvD10fClStKlCvPcG7nNXqnp0X6N3EUKyoa7mbUD\nfww8ABwFHjazo9sc+g13f2vh6+OB65QmUBy5dzdwm71Sh0f7uXR9Xbf+FaGykftx4Ky7n3P3NPBn\nwEPRliXNKB1zuE+P5S+qXphXa0akkrPwIHCh5OeZwmNbvd3MnjazL5vZHUGqk6ZSHLl3xThyB82Y\nEQHoCPR7TgOH3H3ZzB4E/hI4svUgM3sEeATg0KFDgV5akmIz3NtjGrlrrrvIpkrOwovAVMnPk4XH\nNrn7orsvF74/AXSa2djWX+Tuj7n7MXc/Nj4+XkfZkkSpTJaujjbMLJbXH+7rYrCnQ6tURags3J8E\njpjZrWbWBbwXeKL0ADPbb4Uz2syOF37vXOhiJdlSG7nY+u1F02P9vPK6Ru4iZdsy7p4xsw8BXwHa\ngcfd/Vkz+0Dh+UeBdwO/bmYZYA14r7t7hHVLAqWz8Yf7m8YHOHlO4wqRinruhVbLiS2PPVry/aeA\nT4UtTZpNfuQezxz3ottuGeAvvnuR5VSGge5Ql5REmo9WqEowSRm5A7x8dTnWOkTipnCXYFIb2dim\nQRbddks+3M8q3GWXU7hLMEkYuR8e7aOjzTg7q3CX3U3hLsGkNnKxj9w729uYHuvXyF12PYW7BJPK\nZGO/oApw2/iAeu6y6yncJZgktGUg33c/P7+6ea8bkd0o/jNRWkYS2jIAR/YNkM05r+o2BLKLxX8m\nSstIysi9OB1SfXfZzeI/E6VlJGXk/qbxAcwU7rK7xX8mSstIygXV3q52pkb6ePEHS3GXIhIbhbsE\nk84koy0DcHRikOcvL8ZdhkhsknEmSktIZZLRlgE4emCQV+ZWWEll4i5FJBbJOBOl6WVzTibniWjL\nQH7k7g4vqDUju5TCXYJIx7zF3lZHDwwC8JxaM7JLJeNMlKYX9+bYW00M9TDc18lzlxTusjsl40yU\nppfKZAHo7kzGR8rMODoxqJG77FrJOBOl6cW9OfZ2jk4M8sLlRTJZ3YZAdp/knInS1Irh3t2ZjAuq\nAHccHCSVyen2v7IrKdwliGJbJkkj97umRgA4ff5azJWINF5yzkRpapsXVBPSc4f8xh17+7v47msL\ncZci0nDJOROlqaUSNlsG8hdV75oa5rTCXXah5JyJ0tSSGO4Adx8e4eXZFa6vbsRdikhDJetMlKZ1\nY557ci6oAtw1NQzAdy9o9C67i8Jdgti8oJqwkftbpoZpMzj9mi6qyu6SrDNRmlbSVqgW9Xd3cPTA\nIN8+Nxd3KSINlawzUZpWKmH3lil175vGOP3aAqtp3SFSdo/knYnSlJLacwe497YxNrLOd16Zj7sU\nkYZRuEsQm/eWSeDI/Sem99LV3sa3zr4edykiDZO8M1GaUmojuW2Z3q527jk8wje+r3CX3SN5Z6I0\npXQ2R5tBR5vFXcq23vnmW3jhB0ucn1uJuxSRhlC4SxDFLfbMkhnu77pjPwBfefYHMVci0hgKdwki\nvzl28i6mFk3t7eOOA4P8zTMKd9kdFO4SRCqTTeTF1FIP/Nh+Tr92jQvzq3GXIhK5ZJ+N0jRSG7lE\nXkwt9XN3T2IGXzh1Ie5SRCKX7LNRmkYqm0v8yP3gcC/33T7O/37ygnZnkpaX7LNRmkZ+5J7cnnvR\nw8cPcXUpxQn13qXFKdwliHQTjNwB3vnmfdy+b4A/+upLGr1LS0v+2ShNIbWR/AuqAO1txm/+zO2c\nm13hz5+aibsckchUdDaa2f1m9qKZnTWz397meTOzTxSef9rM7g5fqiRZOpv8C6pF77pjP8en9/Kf\n/vp5Ll5bi7sckUiUPRvNrB34Y+AB4CjwsJkd3XLYA8CRwtcjwKcD1ykJl9pI9jz3UmbGf33PW3Dg\nV/7Hk8yvpOMuSSS4SoZax4Gz7n7O3dPAnwEPbTnmIeCznncSGDazicC1SoI1wzz3UodG+3jsl+7h\nlbkV/uUnv8lfPX2J9Y1s3GWJBNNRwTEHgdKJwTPAT1ZwzEHgcl3VbeP/vTTLf/yr50L/WqnTa/Or\n3Dk5HHcZVXn7bWP8+a/9FL/5hTN86H99l852Y3ygm96u9obdRiGZN2uQqP2bn5jiV//pj0T6GpWE\nezBm9gj5tg2HDh2q6XcMdHdwZN9AyLIkgCP7BnjPPZNxl1G1t0wN85WP/DTfenmOk+fmmF1KsZZu\nzAje8Ya8jiTP2EB35K9RSbhfBKZKfp4sPFbtMbj7Y8BjAMeOHavpk33P4RHuOXxPLX9UZFsd7W28\n4/Zx3nH7eNyliARTSZP0SeCImd1qZl3Ae4EnthzzBPC+wqyZtwHX3T14S0ZERCpTduTu7hkz+xDw\nFaAdeNzdnzWzDxSefxQ4ATwInAVWgfdHV7KIiJRTUc/d3U+QD/DSxx4t+d6BD4YtTUREatU8c9dE\nRKRiCncRkRakcBcRaUEKdxGRFqRwFxFpQZaf6BLDC5vNAudr/ONjwOsBy4mT3ksytcp7aZX3AXov\nRYfdveyKu9jCvR5mdsrdj8VdRwh6L8nUKu+lVd4H6L1US20ZEZEWpHAXEWlBzRruj8VdQEB6L8nU\nKu+lVd4H6L1UpSl77iIicnPNOnIXEZGbaNpwN7PfK2zGfcbM/tbMDsRdU63M7L+Y2QuF9/MXZtZc\nWxqVMLP3mNmzZpYzs6ab2VBuM/hmYWaPm9lVM3sm7lrqZWZTZvY1M3uu8Nn6cNw11cLMeszsO2b2\nD4X38R8ifb1mbcuY2aC7Lxa+/w3gqLt/IOayamJm/xz4P4XbK/8BgLv/+5jLqomZvRnIAf8d+Lfu\nfirmkipW2Az+JeBnyG8V+STwsLs33b6OZvbTwDL5vY1/LO566lHYj3nC3U+b2R7gKeBnm+3vxfJ7\nN/a7+7KZdQLfBD5c2Hc6uKYduReDvaAfmnfPMnf/W3fPFH48SX4nq6bk7s+7+4tx11GjSjaDbwru\n/nVgPu46QnD3y+5+uvD9EvA8+T2am4rnLRd+7Cx8RZZbTRvuAGb2+2Z2AfhF4KNx1xPIrwBfjruI\nXWqnjd4lIcxsGrgL+Ha8ldTGzNrN7AxwFfg7d4/sfSQ63M3sq2b2zDZfDwG4+++6+xTwOeBD8VZ7\nc+XeS+GY3wUy5N9PYlXyXkRCM7MB4IvAR7b8y71puHvW3d9K/l/nx80sspZZRTsxxcXd/1mFh36O\n/E5RH4uwnLqUey9m9svAvwDe6Qm/EFLF30uzqWijd2m8Qo/6i8Dn3P1LcddTL3e/ZmZfA+4HIrno\nneiR+82Y2ZGSHx8CXoirlnqZ2f3AvwP+lbuvxl3PLlbJZvDSYIULkZ8Bnnf3P4y7nlqZ2XhxJpyZ\n9ZK/cB9ZbjXzbJkvAj9KfmbGeeAD7t6UoywzOwt0A3OFh0428cyfnwM+CYwD14Az7v6ueKuqnJk9\nCPwRNzaD//2YS6qJmX0euI/83QevAB9z98/EWlSNzOyfAN8Avkf+fAf4ncLezk3DzO4E/if5z1Yb\n8AV3/3hkr9es4S4iIjtr2raMiIjsTOEuItKCFO4iIi1I4S4i0oIU7iIiLUjhLiLSghTuIiItSOEu\nItKC/hGSfaCEhBe8ZAAAAABJRU5ErkJggg==\n", "text/plain": [ "" ] }, "metadata": {}, "output_type": "display_data" }, { "data": { "image/png": "iVBORw0KGgoAAAANSUhEUgAAAXcAAAD8CAYAAACMwORRAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz\nAAALEgAACxIB0t1+/AAAH2BJREFUeJzt3XmQnHd95/H3t+fQ3NNz6RrNaMayJCwfsuSRfITE5jA+\nFixIYtZOdlkTiMsVHEhS7EKFImyWUFmWTSpLYjAOOEDW4JjlsjcCGS/YBoxsjSVZ1kiWrVujczT3\n3dMzv/2je0R7MtK0Rk/308/Tn1eVqtQ9z3R/uzTzqZ++z+8w5xwiIhIuEb8LEBER7yncRURCSOEu\nIhJCCncRkRBSuIuIhJDCXUQkhBTuIiIhpHAXEQkhhbuISAgV+vXG9fX1rqWlxa+3FxEJpJdffvms\nc65hrut8C/eWlhba29v9ensRkUAysyPpXKe2jIhICCncRURCSOEuIhJCCncRkRBSuIuIhJDCXUQk\nhBTuIiIh5Ns8d5GL0T8ywZaOU5wdHuf61jquW17jd0kiOU3hLjnvhQNn+eNv7aB7OHbuuTuvXsz/\nvHstZcX6ERaZjX4zJKftPt7Ph77ezrKaUr523wZa68r53y8e4W+e3kf3UIx//tD1FBequygyk34r\nJGeNxyf52OM7iJYV8dgfXs+1TVGqy4r4yNsu52/ffy0vHurhc/+6x+8yRXKSwl1y1qO/OMyBrmH+\n+revZmFlyZu+9t51jfzBb7TyjV8dYdvhHp8qFMldCnfJScPjcR55/gBvW93ALasXznrNx29bRWO0\nlE99/1Xik1NZrlAktyncJSc90X6M3pEJHnz7yvNeU1ZcyKfffQWvnx7iBztPZLE6kdyncJec45zj\nX7YdY+2y6jmnPN525WLWLKni73/6hkbvIinmDHcze9TMzpjZ7vN83czsi2a238x2mdl678uUfPLq\n8X5eOzXI3W1Nc15rZvzJO1dypHuEf331ZBaqEwmGdEbuXwduv8DX7wBWJv/cD3z50suSfPad9k4W\nFEZ4z9qlaV3/zisW0VJXxjd/ldYZBiJ5Yc5wd849D1xoOsIm4JsuYSsQNbMlXhUo+WVqyrGl4xRv\nW72Q6tKitL4nEjH+ww3LeflIL7uP92e4QpFg8KLn3ggcS3ncmXzu3zCz+82s3czau7q6PHhrCZvd\nJ/o5MzjOrWsWXdT33X1dEyVFEf5Zo3cRIMs3VJ1zjzjn2pxzbQ0Nc57vKnnomT2niRi87S2zT388\nn+qyIjatbeSpXScYicUzVJ1IcHgR7seB1Dtfy5LPiVy0Z/aeoW15LbXlxRf9vb+9vpGR2CRPd5zO\nQGUiweJFuD8JfCA5a+YGoN85p2kLctF6hmPsOTnAb62qn9f3b2ippTFayvd3aGwhMufGYWb2beAW\noN7MOoHPAEUAzrmHgc3AncB+YAT4YKaKlXB76VA3ADeuqJvX90cixqZrl/LwcwfoGhynoXKBl+WJ\nBMqc4e6cu3eOrzvgI55VJHlr68EeSosKuLoxOu/XeN+6Rr707AE2v3qS/3RTi3fFiQSMVqhKzth6\nsJu2lppL2sJ35aJKVjSUs6XjlIeViQSPwl1yQu9wjNdODXLDZfNryaS6/arFvHioh96Uwz1E8o3C\nXXLCzs4+ANY3X/rxebdduZjJKcczezVrRvKXwl1ywq5j/ZjB1cuqL/m1rm6sZml1CVs0JVLymMJd\ncsKuzj5WNFRQseDST340M9515WKef6OL4XEtaJL8pHAX3znn2HW8n2saL33UPu1daxYRi0/xy/1n\nPXtNkSBRuIvvTg2M0TU4zjUetGSmtbXUUl5cwHOvaw8jyU8Kd/HdK8cSOzle0zT/+e0zFRdGuOny\nep7d10ViKYZIflG4i+92dfZRGDHWLKny9HVvXtXA8b5RDnQNe/q6IkGgcBffdZwY4PKFFZQUFXj6\nujevSuw8qtaM5COFu/jujdODvGVxpeev21RbxoqGcp7dd8bz1xbJdQp38dXA2AQn+sdYlYFwB7hl\n9UJePNTDaGwyI68vkqsU7uKrN04PArB6UWbC/eZVDcTiU2xN7jgpki8U7uKrfaeGAFiVoXDf2FpL\ncUGEXx1QuEt+UbiLr14/PUh5cQGN0dKMvH5JUQHrmqO8cECLmSS/KNzFV6+fHmTlokoiEcvYe9y0\nop6OEwP0jWiXSMkfCnfx1eunB1m1qCKj73HT5XU4By8e6sno+4jkEoW7+KZ7aJyzQ7GM9dunrV0W\npbSoQH13ySsKd/HN9MrRyxdmduReXBihraVGfXfJKwp38c3hs4lwv6w+s+EOib7766eH6Bocz/h7\nieQChbv45uDZYYoKjKXRkoy/100rEsf3bT2o1ozkB4W7+Obw2WGaassoLMj8j+GVS6uoXFDIC+q7\nS55QuItvDncP01pXnpX3KiyIsLG1lhc1cpc8oXAXX0xNOQ53D9NSn51wB9jQWsvBs8OcHVLfXcJP\n4S6+ODUwxtjEFK3ZDPeWWgDaD2u+u4Sfwl18MT1TJpvhfnVjNQsKI2w73Ju19xTxi8JdfHGoOxHu\n2WzLFBdGuLYpyjaN3CUPKNzFF4fPDrOgMMKSqsxPg0y1oaWWjhMDDI/Hs/q+ItmWVrib2e1mts/M\n9pvZJ2f5erWZPWVmr5hZh5l90PtSJUwOnR1heV1ZRjcMm82G1lompxw7jvZl9X1Fsm3OcDezAuAh\n4A5gDXCvma2ZcdlHgD3OubXALcDfmFmxx7VKiHT2jtBcW5b1913fHCVi8JJaMxJy6YzcNwL7nXMH\nnXMx4HFg04xrHFBpZgZUAD2A/t8rs3LOcbx3lGU12Q/3ypIirlhSpRkzEnrphHsjcCzlcWfyuVT/\nAFwBnABeBT7mnJvypEIJnYHROIPjcZbVZOaAjrlsaKllx9E+Jib1Iyrh5dUN1duAncBS4FrgH8ys\nauZFZna/mbWbWXtXV5dHby1Bc6x3BMC3cN/YWsvoxCS7j/f78v4i2ZBOuB8HmlIeL0s+l+qDwPdc\nwn7gEPCWmS/knHvEOdfmnGtraGiYb80ScJ29owC+tGUA2lpqAGjXfHcJsXTCfRuw0sxakzdJ7wGe\nnHHNUeAdAGa2CFgNHPSyUAmPTp9H7gsrS2ipK9NNVQm1wrkucM7FzexBYAtQADzqnOswsweSX38Y\n+CzwdTN7FTDgE845nYwgs+rsHaViQSHVpUW+1XDd8lqe3XcG5xyJeQAi4TJnuAM45zYDm2c893DK\n308A7/K2NAmrzt5RltWU+hqq65dH+e72To72jLA8SztTimSTVqhK1nX2jvjWkpm2vjnRd99+VH13\nCSeFu2SVn3PcU61aVEnFgkK2H9FKVQknhbtkld9z3KcVRIy1TdUauUtoKdwlq6bnuDdG/Q13SLRm\nXjs1yEhMi6klfBTuklWn+scAWJoj4T455XjlmBYzSfgo3CWrTg4kwn1JdXa3+p3NuuYooJuqEk4K\nd8mqU/2jFEaMuooFfpdCtKyYyxrK2aFwlxBSuEtWnewfY1FVCQVZ3sf9fNY317D9aB/OOb9LEfGU\nwl2y6lT/GItzoCUzbX1zDT3DMY50j/hdioinFO6SVTkX7svVd5dwUrhL1jjnONk/lvVzUy9k5cLk\nYiaFu4SMwl2yZmA0zujEZE6N3AsixrVNUa1UldBRuEvWnBxI7OO+pNr/Oe6p1jdHee3UAMPjWswk\n4aFwl6w5mVzAlEsjd4B1y2uYcvBKp0bvEh4Kd8ma6dWpubCAKdX6psQOkTuOKtwlPBTukjUn+8eI\nGDRU+r+AKVV1WRErGsrZfkQ3VSU8FO6SNaf6R6mvWEBRQe792K1vrmHHMS1mkvDIvd8yCa2TOTbH\nPdU6LWaSkFG4S9Z0DY6zKIfmuKeaXsy045haMxIOCnfJmq7B8Zzrt09bubCS8uICzXeX0FC4S1ZM\nTE7RMxKjIQd2g5xN4mSmqEbuEhoKd8mKnuEYzuXeTJlU65tr2HtSJzNJOCjcJSvODIwDsDCHw31d\nc5TJKcernTqZSYJP4S5Z0TWUWMCUyyP3dc3JxUzH1HeX4FO4S1Z0DSZG7rkc7rXlxbTUlWkxk4SC\nwl2yYjrc63P0huq0dVrMJCGhcJesODM4TnVpESVFBX6XckHrm6N0DY5zvG/U71JELonCXbIil+e4\np5ruu2/XJmIScAp3yYquwfGcneOeavXiSkqKIuzQyUwScGmFu5ndbmb7zGy/mX3yPNfcYmY7zazD\nzJ7ztkwJujMBGbkXFUS4ZllUI3cJvDnD3cwKgIeAO4A1wL1mtmbGNVHgS8BdzrkrgbszUKsElHMu\nMG0ZSMx333Oin7GJSb9LEZm3dEbuG4H9zrmDzrkY8DiwacY1vwd8zzl3FMA5d8bbMiXIhmOTjE5M\n5vQCplTrm2uYmHR0nBjwuxSReUsn3BuBYymPO5PPpVoF1JjZs2b2spl9YLYXMrP7zazdzNq7urrm\nV7EEThDmuKda15TcIVJ9dwkwr26oFgLXAf8OuA34tJmtmnmRc+4R51ybc66toaHBo7eWXHdmIPdX\np6ZaWFVCY7RUx+5JoBWmcc1xoCnl8bLkc6k6gW7n3DAwbGbPA2uB1z2pUgKtayhYI3eA9ctraD/c\n43cZIvOWzsh9G7DSzFrNrBi4B3hyxjU/BN5qZoVmVgZcD+z1tlQJqum2zMLK3DyoYzbrmqKc7B/j\nZL8WM0kwzRnuzrk48CCwhURgP+Gc6zCzB8zsgeQ1e4EfA7uAl4CvOud2Z65sCZKuwXEKI0a0tMjv\nUtK2fnliMdNOtWYkoNJpy+Cc2wxsnvHcwzMefwH4gnelSVicHRqntryYSMT8LiVta5ZUUVwYYfvR\nXu64eonf5YhcNK1QlYzrGY5RF4DVqamKCyNctbRKN1UlsBTuknHdwzHqyov9LuOirW+uYdfxfmLx\nKb9LEbloCnfJuJ7hGLUBDPd1zTXE4lPsPanFTBI8CnfJuO6hoIa7FjNJcCncJaPG45MMjccD2ZZZ\nGi1lcVWJNhGTQFK4S0b1DMcAqK0IXrhDYvS+45hG7hI8CnfJqO6hRLgHceQOiZuqx3pGzy3EEgkK\nhbtk1LmRe3mwpkJOU99dgkrhLhn163AP5sj9qsZqCiPGjmPqu0uwKNwlo7qT4V4f0J57SVEBVy6t\nYvsRjdwlWBTuklE9w+MURIyqkuDsKzPTuuYadnX2E5/UYiYJDoW7ZFTPcIyasmDtKzPTuuYooxOT\nvHZq0O9SRNKmcJeM6h4K5tYDqdY3J3aIVN9dgkThLhnVHdCtB1ItqymlvqJYM2YkUBTuklE9w7HA\nLmCaZmasa67RDpESKAp3yajuofHAt2Ug0Xc/dHaY3uTsH5Fcp3CXjJmYnGJgLB74tgzAuqbpvrta\nMxIMCnfJmOlRbtAO6pjN2qZqIoZaMxIYCnfJmOkFTGFoy5QVF/KWxTqZSYJD4S4ZE/StB2ZavzzK\nzmN9TE45v0sRmZPCXTImTCN3SPTdh8bj7D8z5HcpInNSuEvGTPfca8IS7skdIrdrvrsEgMJdMqZ3\nJBHu0dLg7iuTqrW+nGhZkRYzSSAo3CVj+kYmqCwppLAgHD9mZsa6pqhuqkoghOO3TnJS30iMaFk4\nRu3T1jXX8MaZIfpHJ/wuReSCFO6SMX2jE9SUhaPfPq1teWIxk/Z3l1yncJeM6R2ZoDok/fZp65pr\nKIwYLx3u8bsUkQtSuEvG9I/EQjdyLy0u4KrGal46pHCX3JZWuJvZ7Wa2z8z2m9knL3DdBjOLm9nv\neleiBFXvyEToeu4A17fWsquzj7GJSb9LETmvOcPdzAqAh4A7gDXAvWa25jzXfR542usiJXgmpxwD\nYxNEQzZyB9jQUsvEpGOnDu+QHJbOyH0jsN85d9A5FwMeBzbNct0fA98FznhYnwTUwOgEzoVnjnuq\ntpbETdVtas1IDksn3BuBYymPO5PPnWNmjcD7gC97V5oEWV9yqmBNefjCPVpWzOpFlbqpKjnNqxuq\nfwd8wjl3wePhzex+M2s3s/auri6P3lpy0a9Xp4avLQOwsbWW7Ud6iU9e8EdexDfphPtxoCnl8bLk\nc6nagMfN7DDwu8CXzOy9M1/IOfeIc67NOdfW0NAwz5IlCPpHEiP3MN5QBdjQWstwbJI9Jwf8LkVk\nVumE+zZgpZm1mlkxcA/wZOoFzrlW51yLc64F+D/AHznnfuB5tRIY50buIbyhCrCxpRZAUyIlZ80Z\n7s65OPAgsAXYCzzhnOswswfM7IFMFyjB1JccudeEdOS+uLqEptpStqnvLjmqMJ2LnHObgc0znnv4\nPNfed+llSdD1jcQwg8qScIY7JKZEPruvC+ccZuZ3OSJvohWqkhF9o4mtBwoi4Q2961tr6RmOcaBL\nh3dI7lG4S0b0joRv07CZNpzru2sTMck9CnfJiL6RWOg2DZuptb6c+opi9d0lJyncJSP6RiZCezN1\nmplxfWsdWw9245wOzZbconCXjOgdiYV2GmSqG1fUcbJ/jMPdI36XIvImCnfJiP6Q7gg5000r6gB4\n4cBZnysReTOFu3huYnKKwfF4aLceSNVaX87iqhJeONDtdykib6JwF8/1h3jTsJnMjJtW1LH1QDdT\nU+q7S+5QuIvn+pJbD4R9tsy0my6vp3s4xr7Tg36XInKOwl0813du07Dwt2UgcVMVUGtGcorCXTw3\nMJYI93wZuTdGS2mpK+NXuqkqOUThLp6b7rnnS7gD3LiinhcP9mh/d8kZCnfx3MBoHICqkrT2pQuF\nm1bUMTgeZ/cJ7e8uuUHhLp6bHrlX5dXIPdF3/+V+tWYkNyjcxXMDoxOUFRdQVJA/P171FQt4y+JK\nLWaSnJE/v32SNQNjE1SFeB/38/mNy+vZdriX0dik36WIKNzFe/3Jvdzzzc2rGojFp9h6SFMixX8K\nd/HcwGicqtL8uZk6bWNrLSVFEZ7b1+V3KSIKd/Fevo7cS4oKuPGyOp57XeEu/lO4i+fytecOidbM\nobPDHOke9rsUyXMKd/HcwOhEXk2DTHXz6oUAPK/Ru/hM4S6emppyDI7H8zbcW+rKaK4tU2tGfKdw\nF08NjsdxLr9Wp6YyM25e1cALB7oZj2tKpPhH4S6eGsjDfWVmunlVAyOxSV4+3Ot3KZLHFO7iqXzc\nemCmG1fUUVwQ4Wf7zvhdiuQxhbt4anq733ydLQNQvqCQG1bU8czeMzin05nEHwp38ZTaMgm3rlnE\nobPDHOga8rsUyVMKd/HUue1+83CFaqpbr1gEwJaO0z5XIvlK4S6eyseDOmazuLqEtcuq+ckehbv4\nI61wN7PbzWyfme03s0/O8vXfN7NdZvaqmb1gZmu9L1WCYGBsgohBeXF+j9wh0ZrZeayPMwNjfpci\neWjOcDezAuAh4A5gDXCvma2Zcdkh4Gbn3NXAZ4FHvC5UgmFgdILKkiIiEfO7FN/dumYxAD/Zq9G7\nZF86I/eNwH7n3EHnXAx4HNiUeoFz7gXn3PSk3q3AMm/LlKDI103DZrNqUQXNtWVqzYgv0gn3RuBY\nyuPO5HPn8yHgR7N9wczuN7N2M2vv6tLy7DAaGMvP7X5nY2a8a80iXtjffW6KqEi2eHpD1czeRiLc\nPzHb151zjzjn2pxzbQ0NDV6+teQIjdzf7M5rlhCbnOInmjUjWZZOuB8HmlIeL0s+9yZmdg3wVWCT\nc05H0eSpgdH83e53NuuaojRGS3lq1wm/S5E8k064bwNWmlmrmRUD9wBPpl5gZs3A94D/6Jx73fsy\nJSgGxjRyT2VmvGftUn7xxll6hmN+lyN5ZM5wd87FgQeBLcBe4AnnXIeZPWBmDyQv+wugDviSme00\ns/aMVSw5rT+P93I/n/esXUJ8yvHj3af8LkXySFp3vpxzm4HNM557OOXvHwY+7G1pEjSx+BRjE1N5\nu93v+axZUsVl9eU89coJfu/6Zr/LkTyhFarimXObhmnk/iZmxrvXLmXroW4taJKsUbiLZ4bGEvvK\nVCzQyH2mu9YuwTl4atdJv0uRPKFwF88MJsO9UrNl/o3LF1ZyzbJqvtN+TNsAS1Yo3MUzg8m2TKV6\n7rO6u62J104N0nFiwO9SJA8o3MUzg+Nqy1zIXWuXsqAwwhPtx+a+WOQSKdzFM9NtGS1iml11aRG3\nX7WYH+w4ztiEDs+WzFK4i2fUlpnb+9uaGBiLs6VDc94lsxTu4plzs2UU7ud142V1NNWW8tiLR/0u\nRUJO4S6eGRyPU1IUoahAP1bnE4kYH7ihhZcO9dBxot/vciTE9Fsonhkci1OxQP32uby/rYnSogK+\n/svDfpciIaZwF88Mjk1o64E0VJcV8TvXNfLDV07QPTTudzkSUgp38czgWFw3U9N0300txOJTfEu9\nd8kQhbt4Zmg8rpupabp8YSU3r2rgG786zGhM0yLFewp38czg2ASV6rmn7cG3X87ZoRjffkmjd/Ge\nwl08o7bMxdnQUssNl9XylecPaFGTeE7hLp4ZGlNb5mJ99O0rOT0wri0JxHMKd/HE1JRjKBbXjpAX\n6cYVdWxsreWL/+8NhpJ784h4QeEunhiKxXEOTYW8SGbGp+68grNDMb7y3AG/y5EQUbiLJ3RQx/yt\nbYpy19ql/OPPD3Kib9TvciQkFO7iCR3UcWn+822rcQ7+8qkOv0uRkFC4iyemd4TUDdX5aaot409v\nXcWWjtP8eLeO4pNLp3AXT0wf1KGpkPP34be2cuXSKj79ww76RmJ+lyMBp3AXT/z6oA6F+3wVFkT4\n/O9cQ99IjI9/5xWdtSqXROEunjjXltEK1UtyVWM1f37nFTyz9wxf/fkhv8uRAFO4iyeGxtSW8cp9\nN7Vwx1WL+e8/fo2fvnba73IkoBTu4onBsTgRg7LiAr9LCTwz4wt3r2XNkir+6LHtvHyk1++SJIAU\n7uKJofE4FQsKMTO/SwmFigWF/NMHN7C4qoT7Hn2JFw6c9bskCRiFu3hiYGxCc9w9Vl+xgG/94Q0s\nri7hvke38f0dnX6XJAGSVrib2e1mts/M9pvZJ2f5upnZF5Nf32Vm670vVXKZdoTMjKXRUr7zwI1c\n2xzlT//lFf7siZ30DGuapMxtznA3swLgIeAOYA1wr5mtmXHZHcDK5J/7gS97XKfkuCGFe8ZEy4r5\n1oev56PvWMkPd57g5v/xMx762X6FvFxQOr+NG4H9zrmDAGb2OLAJ2JNyzSbgmy4xMXermUXNbIlz\nTkvt8sTg+AQLK0v8LiO0Cgsi/Nmtq3jPNUv46x+9xhe27ON/PfMGt6xu4DdX1rO2KUpLfTlVao1J\nUjrh3gikbjbdCVyfxjWNgOfh/tzrXfzV/90z94WSVUe6R7isvsLvMkJv5aJKHr1vA/tODfLtl47y\nkz2neXrPr6dLlhcXULagkLLiAgojs9/cPt9Nb90Kz55/v6GJD//mZRl9j6z+P9rM7ifRtqG5uXle\nr1GxoJCVixQiuWblogru2dDkdxl5Y/XiSv7rXVfymfes4WjPCHtPDnK4e5iuwXFGYpMMj8eZnG2F\n63kWvbrzfUEyor5iQcbfI51wPw6k/tYuSz53sdfgnHsEeASgra1tXj9N1y2v4brl183nW0VCx8xY\nXlfO8rpyv0uRHJPObJltwEozazWzYuAe4MkZ1zwJfCA5a+YGoF/9dhER/8w5cnfOxc3sQWALUAA8\n6pzrMLMHkl9/GNgM3AnsB0aAD2auZBERmUtaPXfn3GYSAZ763MMpf3fAR7wtTURE5ksrVEVEQkjh\nLiISQgp3EZEQUriLiISQwl1EJITMr3MazawLODLPb68HwrLBtT5LbgrLZwnL5wB9lmnLnXMNc13k\nW7hfCjNrd861+V2HF/RZclNYPktYPgfos1wstWVEREJI4S4iEkJBDfdH/C7AQ/osuSksnyUsnwP0\nWS5KIHvuIiJyYUEduYuIyAUENtzN7LPJw7h3mtnTZrbU75rmy8y+YGavJT/P980s6ndN82Vmd5tZ\nh5lNmVngZjbMdRh8UJjZo2Z2xsx2+13LpTKzJjP7mZntSf5sfczvmubDzErM7CUzeyX5Of4yo+8X\n1LaMmVU55waSf/8osMY594DPZc2Lmb0L+Glye+XPAzjnPuFzWfNiZlcAU8BXgI8759p9LiltycPg\nXwduJXFU5DbgXudc4M51NLPfAoZInG18ld/1XAozWwIscc5tN7NK4GXgvUH7d7HE+YblzrkhMysC\nfgF8zDm3NRPvF9iR+3SwJ5Vz3gPEcp9z7mnnXDz5cCuJk6wCyTm31zm3z+865uncYfDOuRgwfRh8\n4Djnngd6/K7DC865k8657cm/DwJ7SZzRHCguYSj5sCj5J2O5FdhwBzCzz5nZMeD3gb/wux6P/AHw\nI7+LyFPnO+hdcoSZtQDrgBf9rWR+zKzAzHYCZ4CfOOcy9jlyOtzN7Bkz2z3Ln00AzrlPOeeagMeA\nB/2t9sLm+izJaz4FxEl8npyVzmcR8ZqZVQDfBf5kxv/cA8M5N+mcu5bE/843mlnGWmZpncTkF+fc\nO9O89DESJ0V9JoPlXJK5PouZ3Qe8G3iHy/EbIRfx7xI0aR30LtmX7FF/F3jMOfc9v+u5VM65PjP7\nGXA7kJGb3jk9cr8QM1uZ8nAT8JpftVwqM7sd+C/AXc65Eb/ryWPpHAYvWZa8Efk1YK9z7m/9rme+\nzKxheiacmZWSuHGfsdwK8myZ7wKrSczMOAI84JwL5CjLzPYDC4Du5FNbAzzz533A3wMNQB+w0zl3\nm79Vpc/M7gT+jl8fBv85n0uaFzP7NnALid0HTwOfcc59zdei5snM3gr8HHiVxO87wJ8nz3YODDO7\nBvgGiZ+tCPCEc+6/Zez9ghruIiJyfoFty4iIyPkp3EVEQkjhLiISQgp3EZEQUriLiISQwl1EJIQU\n7iIiIaRwFxEJof8PxhrwW8mFKH0AAAAASUVORK5CYII=\n", "text/plain": [ "" ] }, "metadata": {}, "output_type": "display_data" } ], "source": [ "mybeta = st.beta(2,4) # default location and scale\n", "r = np.linspace(-3,3,1000)\n", "plt.plot(r,mybeta.pdf(r))\n", "mybeta1 = st.beta(2,4,-1) # shift 1 to the left\n", "plt.figure()\n", "plt.plot(r,mybeta1.pdf(r))\n", "mybeta2 = st.beta(2,4,-1,2) # shift 1 to the left and scale by 2\n", "plt.figure()\n", "plt.plot(r,mybeta2.pdf(r))" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "In Scipy, the normal random variable is defined using its location and scale as st.norm(loc,scale). The location represents the mean and the scale represents the standard deviation:" ] }, { "cell_type": "code", "execution_count": 3, "metadata": {}, "outputs": [ { "data": { "text/plain": [ "" ] }, "execution_count": 3, "metadata": {}, "output_type": "execute_result" }, { "data": { "image/png": "iVBORw0KGgoAAAANSUhEUgAAAZIAAAEWCAYAAABMoxE0AAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nM/8Frir1AAAAAElFTkSuQmCC\n", "text/plain": [ "" ] }, "metadata": {}, "output_type": "display_data" }, { "data": { "image/png": "iVBORw0KGgoAAAANSUhEUgAAAYUAAAEWCAYAAACJ0YulAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n", "text/plain": [ "" ] }, "metadata": {}, "output_type": "display_data" } ], "source": [ "mynorm = st.norm(0,2) # normal RV with mean 0 and variance 4 (standard deviation 2)\n", "r = np.linspace(-10,10,1000)\n", "plt.plot(r,mynorm.pdf(r))\n", "plt.xlabel('x')\n", "plt.ylabel('f(x)')\n", "plt.title('pdf of a N(0,4)')\n", "plt.figure()\n", "plt.plot(r,mynorm.cdf(r))\n", "plt.xlabel('x')\n", "plt.ylabel('F(x)')\n", "plt.title('CDF of a N(0,4)')" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "#### _$\\color{red}{Exercise\\ 1}$_: \n", "Plot the pdf and CDF for exponential and Gamma distributions. Use [st.expon(location,scale)](https://docs.scipy.org/doc/scipy/reference/generated/scipy.stats.expon.html), [st.gamma(a,loc,scale)](https://docs.scipy.org/doc/scipy/reference/generated/scipy.stats.gamma.html). Pick the parameters arbitrarily." ] }, { "cell_type": "code", "execution_count": 4, "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "Exponential with rate parameter 3, Exp(3), (mean and standard deviation 1/3)\n", "Gamma which has distn equal to that of Exp(3)+Exp(3)\n" ] }, { "data": { "image/png": "iVBORw0KGgoAAAANSUhEUgAAAXcAAAEICAYAAACktLTqAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n4S4iMgQp3EVEhiCFu4jIEKRwFxEZgv4/OV4wqYgEQ08AAAAASUVORK5CYII=\n", "text/plain": [ "" ] }, "metadata": {}, "output_type": "display_data" }, { "data": { "image/png": "iVBORw0KGgoAAAANSUhEUgAAAXcAAAEICAYAAACktLTqAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nEZFZ6P8DwMX21jq2KYQAAAAASUVORK5CYII=\n", "text/plain": [ "" ] }, "metadata": {}, "output_type": "display_data" } ], "source": [ "r = np.linspace(-10,10,1000)\n", "print(\"Exponential with rate parameter 3, Exp(3), (mean and standard deviation 1/3)\")\n", "print(\"Gamma which has distn equal to that of Exp(3)+Exp(3)\")\n", "myexp = st.expon(0,1/3)\n", "plt.plot(r,myexp.pdf(r));\n", "plt.plot(r,myexp.cdf(r));\n", "plt.legend([\"pdf\",\"cdf\"],loc=2);\n", "plt.title(\"Exp(2); rate = 2\");\n", "plt.figure()\n", "mygamma = st.gamma(2,0,1/3)\n", "plt.plot(r,mygamma.pdf(r));\n", "plt.plot(r,mygamma.cdf(r));\n", "plt.title(\"Gamma(2,2); shape = 2, rate = 2\");\n", "plt.legend([\"pdf\",\"cdf\"],loc=2);\n", "\n" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "## Sampling from distributions\n", "Law of large numbers (LLN) states that under certain conditions the average of many samples of a random variable goes to its expected value as the number of samples increases. \n", "$$\\frac1n\\sum_{i=1}^n X_i\\to E[X] \\text{ as } n\\to \\infty$$\n", "This allows us to estimate the mean and variance of a RV by taking many samples. For variance, in the above equation, we replace $X$ with $(X-E[X])^2$. An example is shown for $N(0,4)$." ] }, { "cell_type": "code", "execution_count": 5, "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "0.00602047157063\n", "3.98614364417\n" ] } ], "source": [ "X = st.norm.rvs(0,2,size=10000) # Getting 10000 samples from N(0,4)\n", "m = sum(X)/len(X) # Average as an estimate of mean\n", "print(m) # Compare with 0\n", "v = sum([(x-m)*(x-m) for x in X])/len(X)\n", "print(v) # Compare with 4" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "#### _$\\color{red}{Exercise\\ 2}$_: \n", "Pick arbitrary but non-trivial distributions for X and Y. By generating samples, estimate $E[X+Y^2]$, $EX+(EY)^2$, $EX+(EY^2),$ and compare the estimates." ] }, { "cell_type": "code", "execution_count": 6, "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "E[X+Y^2]: 35.0272236526\n", "E[X]+E[Y]^2: 9.76998486677\n", "E[X]+E[Y^2]: 35.0272236526\n", "By the linearity of expectation, we expect the first and third values to be the same!\n" ] } ], "source": [ "X = st.norm.rvs(0,2,size=10000) # Getting 10000 samples from N(0,4)\n", "Y = st.norm.rvs(3,5,size=10000) # Getting 10000 samples from N(3,25)\n", "Y2 = [Y[i]*Y[i] for i in range(10000)]\n", "XY2 = [X[i]+Y[i]*Y[i] for i in range(10000)]\n", "EX = sum(X)/len(X) # Average as an estimate of mean\n", "EY = sum(Y)/len(Y) # Average as an estimate of mean\n", "EY2 = sum(Y2)/len(Y2) # Average as an estimate of mean\n", "EXY2 = sum(XY2)/len(XY2) # Average as an estimate of mean\n", "print(\"E[X+Y^2]:\",EXY2)\n", "print(\"E[X]+E[Y]^2:\",EX+EY*EY)\n", "print(\"E[X]+E[Y^2]:\",EX+EY2)\n", "print(\"By the linearity of expectation, we expect the first and third values to be the same!\")" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "## Samples and histogram\n", "If we take many samples from a given distribution, the histogram of those samples will be similar to the pdf because the fraction of samples in each small interval around a given value tends to the value of the pdf at that point when the number of samples is large. We can see this in action. (Natuarally, this can be used to estimate the pdf from samples. The method is called [Kernel Density Estimation](https://en.wikipedia.org/wiki/Kernel_density_estimation).)" ] }, { "cell_type": "code", "execution_count": 7, "metadata": {}, "outputs": [ { "data": { "text/plain": [ "[]" ] }, "execution_count": 7, "metadata": {}, "output_type": "execute_result" }, { "data": { "image/png": "iVBORw0KGgoAAAANSUhEUgAAAYQAAAD8CAYAAAB3u9PLAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz\nAAALEgAACxIB0t1+/AAAIABJREFUeJzt3Xt8VOW18PHfSiDcrxK5JRDACEZRwICggChoARX0fKzC\naa0e21LeitYebY3tec9r6+k52NbaemqlXktbK9VWEQRBQEGQW8LdiEi4ByHcCdeEJOv9Y+/AOOay\nk8zMnsv6fj7zmX159uw1e5JZs5/n2c8WVcUYY4xJ8jsAY4wx0cESgjHGGMASgjHGGJclBGOMMYAl\nBGOMMS5LCMYYYwBLCMYYY1yWEIwxxgCWEIwxxrga+R1AXXTo0EEzMjL8DsMYY2LKmjVrDqlqam3l\nYiohZGRkkJeX53cYxhgTU0Rkl5dyVmVkjDEGsIRgjDHGZQnBGGMMYAnBGGOMyxKCMcYYwBKCMcYY\nlyUEY4wxgCUEY4wxLksIxhhjgBi7UtmYaJGRM+cry3ZOvcWHSIwJHTtDMMYYA3hMCCIyWkS2iEiB\niORUsf4bIrJRRDaJyHIRuaq2bUWkvYgsEJGt7nO70LwlY4wx9VFrQhCRZOA5YAyQBUwUkaygYjuA\n61W1L/Ak8IKHbXOARaqaCSxy540xxvjESxvCIKBAVbcDiMgMYDzwaWUBVV0eUH4lkOZh2/HACLfc\ndGAx8Fg934cxIRPcPlBr24AqKeVlYYzImMjwkhC6AnsC5guBa2oo/23gPQ/bdlTVfe70fqBjVS8m\nIpOASQDdunXzEK4xEaDK9TvW8u3cmQws/JRmZSXwl4vh1lvh8cfhkkv8jtCYOgtpo7KI3ICTEOr0\nS19VFdBq1r2gqtmqmp2aWuv9HYwJv+PH+cPM/2H6m/+P4TvX0ayshDJJggMH4JVXICsLfvUr0Cr/\npI2JWl4Swl4gPWA+zV32JSJyJfASMF5VD3vYtkhEOrvbdgYO1C10Y3xw6BAMG8bYz5dTnNKc/xlx\nHwMefI1LfvQO5OfDfffBuXPw4x/D5MmWFExM8ZIQcoFMEekhIinABGBWYAER6Qa8Bdyjqp973HYW\ncK87fS/wTv3fhjERcOoU3HwzbNpEQfs0br3vd/zxmjs50rwNiDhnBq++Cv/8JzRrBi+8AI9Zs5iJ\nHbUmBFUtA6YA84HNwBuqmi8ik0VkslvsP4GLgD+IyHoRyatpW3ebqcBNIrIVGOXOGxOdVGHSJFi3\nDnr1YuLE/2Z3u85Vl/2Xf4G33oLGjZ2qo7//PbKxGlNPojF0Spudna12T2UTblX2MvrLX+Bb34KW\nLWHVKjL+vKPW19mZvhOmTHG22bgRevQIU8TG1ExE1qhqdm3l7EplY2pTVAQPP+xMP/usUzXkxfe/\n75wtnDzpJIYY+vFlEpMlBGNq88gjcOQI3HST02jslQg89xy0aQNz58I//hG2EI0JBUsIxtSg776t\n8Npr0KQJ/PGPzpd8XXTqBFPd5rEf/QhKSkIfpDEhYgnBmOqo8vjiV53phx6qfxvAd78Ll18Ou3Y5\nPY+MiVKWEIypxrW7NnDt7o3Qrp1z9XF9JSfDL37hTD/5pNOmYEwUsvshGFON/7PSrfN/5BEnKdTR\nl3oraTJvdenNgC+2wEsvXWikNiaK2BmCMVW4Yn8Bw3at52RKM6e3UEOJ8IfBdznTzzwDZTYYnok+\nlhCMqcJk9+zgb1eNrtfZQVUWXTIQeveG3butx5GJSlZlZEyQrscPMObz5ZQmNeLlgeP57ypul1kf\nKklO9dOkSc4VzHffXfdeS8aEkZ0hGBNkwob5JGsF7/W+jqJWHUL74vfcA6mpsHYtrFwZ2tc2poHs\nDMEktOBhKhqVl7Fi43wA/tp/TOh32LQp3H8/PPWU0wV1yJDQ78OYerIzBGMC3LR1JamnjvH5Rd3I\nTbs8PDv5znec57//HY4dC88+jKkHSwjGBPjX9fMAeK3/mPDV719yCYwcCWfOOFdBGxMlLCEY4+pc\nfJDrdm3gbKMU3r78hrDsIyNnDhk5c3ighTPw5Kc/fzos+zGmPiwhGOMa/+kSklAWXnINxU1bhnVf\nCzIHc6xpS7IO7IBPPgnrvozxyhKCMQCq3JH/AQBvXz4i7LsrbdSYub2HOjNWbWSihKeEICKjRWSL\niBSISE4V6/uIyAoRKRGRRwOW93bvoFb5KBaRh911T4jI3oB1Y0P3toypm8sO7qD3od0cadaaj3oM\niMg+Z1Ymntdfh4qKiOzTmJrU2u1URJKB54CbgEIgV0RmqeqnAcWOAA8Btwduq6pbgH4Br7MXeDug\nyDOq+usGvQNjQuD2/MUAvNtnGOeSG0dkn7lpWextlUrXXbtg+XIYOjQi+zWmOl7OEAYBBaq6XVVL\ngRnA+MACqnpAVXOBczW8zkhgm6ruqne0xoRBUkU54z9dDAT8ao8AlSRmZw0H4K8P/vf5Bmdj/OIl\nIXQF9gTMF7rL6moC8HrQsgdFZKOIvCIioRkwxpg6GvDFZ3Q6eYTdbTqytkufiO57ZtYIAG75bBmN\nym3AO+OviDQqi0gKMA54M2Dx80BPnCqlfUCV/e9EZJKI5IlI3sGDB8Meq0k8Y7YsB+C93tdFfGyh\nzy7uwecXdaPd2RNcs8d6Gxl/eUkIe4H0gPk0d1ldjAHWqmpR5QJVLVLVclWtAF7EqZr6ClV9QVWz\nVTU7NTW1jrs1phaqfO1zJyHMv9SfYSQq9zvajcMYv3hJCLlApoj0cH/pTwBm1XE/EwmqLhKRzgGz\ndwD288hEXN/9BaQVH2R/y/as69LblxjmuQnh5q0rEbXeRsY/tfYyUtUyEZkCzAeSgVdUNV9EJrvr\np4lIJyAPaA1UuF1Ls1S1WERa4PRQ+l7QS/9SRPoBCuysYr0xYTfm848B51e6ij+X5eR37EVh64tJ\nKz5A/y+2ALf5EocxnkY7VdW5wNygZdMCpvfjVCVVte0p4KIqlt9Tp0iNCTVVRrvtB/Muvc6/OER4\nP3Mw96+Zxc2fr/AvDpPw7Eplk7jy8+l59AuONGvN6vQwjWzq0bze1wIw+vMVoOprLCZxWUIwiett\n5xrJBZdcQ3lSsq+h5HW9jEPN25BxbJ+NbWR8YwnBJK533wWcgeb8VpGUzIJLrnFmZs70NxiTsCwh\nmMR04ADk5lKS3JiPu1/ldzQALKpMCHPsamXjD0sIJjG99x6osrJbX86kNPU7GgCWd7+SkuRGsHo1\n2EWYxgd2T2WTMALHCfr9zJe4Ffig10D/AgpyOqUZq9L7MnznOpg3D+6xjngmsuwMwSScRuVlDN+x\nFoiuhADwYS/nTmrMnVtzQWPCwBKCSTjZez+ldelptl6Uzp62nfwO50vOJ6h586DMBrszkWUJwSSc\nG7blAdF3dgCwq10XyMyEY8dg5Uq/wzEJxhKCSTg3bssFAqpnos0ttzjP1tvIRJglBJNQ0o7tJ/Pw\nHoqbtCCva5bf4VRtrHs3WWtHMBFmCcEklOE71wGwrPtVlCVHaSe74cOhRQvYuBH27Km9vDEhYgnB\nJJRhO5yEsLTHAJ8jqUGTJjBypDO9YIG/sZiEYgnBJIzkinKu3bUBgKUZ/XyOphY33eQ8W0IwEWQJ\nwSSMK/dtpU3JKXa060xhlHU3/YrKhLBwIVTYTXNMZFhCMAljaGX7QUZ/nyPx4NJLIT0dDh2CDRv8\njsYkCE8JQURGi8gWESkQkZwq1vcRkRUiUiIijwat2ykim0RkvYjkBSxvLyILRGSr+9yu4W/HmOoN\ncxPC0lhICCJWbWQirtaEICLJwHPAGCALmCgiwf31jgAPAb+u5mVuUNV+qhrY8TsHWKSqmcAid96Y\n8Cgupv8XWyiTJFZ0v9LvaLyxhGAizMsZwiCgQFW3q2opMAMYH1hAVQ+oai5wrg77Hg9Md6enA7fX\nYVtj6mbxYhpXlLO+S29ONGnhdzTeVPY0WroUzpzxNxaTELwkhK5AYGfoQneZVwosFJE1IjIpYHlH\nVd3nTu8HOtbhNY2pG/dX9rJo710UKDUV+veHkhJYtszvaEwCiESj8lBV7YdT5fSAiAwPLqCqipM4\nvkJEJolInojkHbQx4k19vf8+ECPtB4Gs2shEkJeEsBdID5hPc5d5oqp73ecDwNs4VVAARSLSGcB9\nPlDN9i+oaraqZqempnrdrTEX7NoFn39OcUpzNnS+1O9o6sYSgokgLwkhF8gUkR4ikgJMAGZ5eXER\naSEirSqngZuByjuIzwLudafvBd6pS+DGeOZ+ma7sfmX0DldRnaFDoWlTWL/eue2nMWFUa0JQ1TJg\nCjAf2Ay8oar5IjJZRCYDiEgnESkE/h34DxEpFJHWOO0Cy0RkA7AamKOq89yXngrcJCJbgVHuvDGh\nt2gR4IxfFHOaNoVhw5xp930YEy6efi6p6lxgbtCyaQHT+3GqkoIVA1X+F6rqYWCk50iNqQ9V+PBD\nAJbHYkIAp9powQLnquWJE/2OxsQxu1LZxLfPPoOiIujYkYKL0msvH41uvNF5dhObMeESYxWqxtRR\n5ZfoiBHO1b8xICPnyzfGSaooZ3vbtrBjh9NA3r27T5GZeGdnCCa+VSaEG27wN44GqEhKZkGH3gA8\n+v3fkpEz5ytJw5hQsIRg4ldFBSxe7EzHcEIAWO4OtzFklw10Z8LHEoKJX/n5zmihXdwb18ewFd3c\nhLB7k9NQbkwYWEIw8Svw7CBG2g+qsyW1O0eatabLiUN0P7av9g2MqQdLCCZ+xUH7QSWVJFamXwHA\nkF0bfY7GxCtLCCY+VVTAkiXOdBwkBOD8sN1Ddm/yORITrywhmPi0cSMcOQLdukGPHn5HExLLuzkX\n1l27e4O1I5iwsOsQTHxyq4v+0TqTRx+fW0vh2LDtojQOtGjHxaeO0utwod/hmDhkZwgmPrkJYWW3\nvj4HEkIi59/PkN3WjmBCzxKCiT/l5fDRR8CF7prx4kL3U0sIJvQsIZj4s24dHD/O7jYd2dvmYr+j\nCakV7hnC4N2bnIZzY0LIEoKJP+71B/F2dgCws10XvmjVgYvOFMMnn9S+gTF1YAnBxB+3/aCym2Zc\nETl/lmCjn5pQs4Rg4ktZGSxdClyoXok3KyvPfCwhmBDzlBBEZLSIbBGRAhHJqWJ9HxFZISIlIvJo\nwPJ0EflQRD4VkXwR+UHAuidEZK+IrHcfY0PzlkxCW7MGTpyAzEyKWnXwO5qwqDzzOT5vIT1/PMtG\nPzUhU2tCEJFk4DlgDJAFTBSRrKBiR4CHgF8HLS8DHlHVLGAw8EDQts+oaj/3ER+dxY2/4mi4iuoU\ntunInjYdaVNyissO7PA7HBNHvJwhDAIKVHW7qpYCM4DxgQVU9YCq5gLngpbvU9W17vQJnHsydw1J\n5MZUJQESAlyoDrPupyaUvCSErsCegPlC6vGlLiIZQH9gVcDiB0Vko4i8IiLt6vqaxnxJaSksW+ZM\njxjhayjh9qXhsI0JkYg0KotIS+CfwMOqWuwufh7oCfQD9gFPV7PtJBHJE5G8gwcPRiJcE6tyc+H0\nabjsMujUye9owqoyIQza8wnJFeU+R2PihZeEsBcIvDt5mrvMExFpjJMMXlPVtyqXq2qRqparagXw\nIk7V1Feo6guqmq2q2ampqV53axJR5f0P4vzsAGB/6w7saNeZVqVnuGJ/gd/hmDjhJSHkApki0kNE\nUoAJwCwvLy4iArwMbFbV3wSt6xwwewdgV9mYhkmQ9oNKVm1kQq3WhKCqZcAUYD5Oo/AbqpovIpNF\nZDKAiHQSkULg34H/EJFCEWkNXAfcA9xYRffSX4rIJhHZCNwA/DD0b88kjJIS+PhjZzoBzhDgwvUI\n1rBsQsXT8Ndul9C5QcumBUzvx6lKCrYMqPLehap6j/cwjanFqlVw9ixccQUkSNViZU+j7MJPaVRe\n5nM0Jh7YlcomPiRYdRHAwZbt2dY+jRbnznLlvq1+h2PigCUEEx8SMCFAwOine6wdwTScJQQT+86c\ngRUrQASGD/c7mog637C8y9oRTMPZLTRN7FuxAkpLyb+4J7f8aqXf0URU5R3UsvdudhrWmzTxOSIT\ny+wMwcS+8/c/iM/RTWtyuEVbtnToRrOyEli92u9wTIyzhGBi3/n7J8fh/Q88WGHDYZsQsYRgYtvp\n07BqFeWSxOr0y/2OxheWEEyoWEIwse3jj+HcOfI79qS4aUu/o/HFqm5XUIE4bSlnz/odjolhlhBM\nbKu8XWaCVhcBHGvWms8uznAalVes8DscE8MsIZjYdj4hJF6DcqCV6XafZdNwlhBM7DpxwhnyOjmZ\n3LTEbD+oVHlbTUsIpiEsIZjYtWwZlJdDdjanmjT3OxpfrUq/wrkwb9Uqp6HdmHqwhGBiV+Wv4QQZ\n3bQmxU1bQv/+cO4cLF/udzgmRllCMLGr8oY4CTZ+UbUqj4NVG5l6soRgYk5Gzhz6/vANyvPWcC4p\nmaz5J/wOKTpYQjANZAnBxKSBe/JJ1go2dL6U0ynN/A4nOgwbBsnJTkP7yZN+R2NikKeEICKjRWSL\niBSISE4V6/uIyAoRKRGRR71sKyLtRWSBiGx1n9s1/O2YRFF5l7BEvv7gK1q3hquvhrIyp8HdmDqq\nNSGISDLwHDAGyAImikhWULEjwEPAr+uwbQ6wSFUzgUXuvDGeVN5HONGvPwiUkTOHaZIOwLQnXiIj\nZ47PEZlY4+UMYRBQoKrbVbUUmAGMDyygqgdUNRc4V4dtxwPT3enpwO31fA8mwbQ5c4Ksou2UJjVi\nbdc+focTVSrPmAbbfZZNPXhJCF2BPQHzhe4yL2ratqOq7nOn9wMdPb6mSXDX7PmEJJR1XXpztnFT\nv8OJKrlpWZxLSqbv/m20KjnldzgmxkRFo7KqKqBVrRORSSKSJyJ5Bw8ejHBkJhpVth8k6nDXNTmd\n0oyNnTJJ1goG7sn3OxwTY7wkhL1AesB8mrvMi5q2LRKRzgDu84GqXkBVX1DVbFXNTk1N9bhbE88G\nV7YfdLf2g6pUDmMxxKqNTB15SQi5QKaI9BCRFGACMMvj69e07SzgXnf6XuAd72GbhHXoEJcd3ElJ\ncmPWdbH2g6qcv8+ymziN8arWeyqrapmITAHmA8nAK6qaLyKT3fXTRKQTkAe0BipE5GEgS1WLq9rW\nfempwBsi8m1gF3BXqN+ciUMffAA4deUljVJ8DiY6renah9KkRmQVbYcjR6B9e79DMjGi1oQAoKpz\ngblBy6YFTO/HqQ7ytK27/DAwsi7BGsPChQB8nNHP50Ci19nGTVnXpTfXFObD0qUwfnztGxlDlDQq\nG+OJKixYAMCy7pYQarLSbqtp6sESgokd27fDzp0ca9qS/I49/Y4mqq3sZjfMMXVnCcHEDre6aHm3\nK6lISvY5mOi2tmsfzjZKgY0b4UCVHfiM+QpLCCZ2WPuBZyWNUlhdeRc597gZUxtLCCY2lJef72G0\nNKO/z8HEhmWVidNtdzGmNpYQTGxYv97pQpmRwe62nfyOJiYs7eEmzgULnAZ5Y2phCcHEhspqj1Gj\nnHsHm1p9lpoBF18Me/fC5s1+h2NigCUEE/Uycuaw9PkZAEw5YBdZeaWS5CRQsGoj44klBBP1mpwr\nYVChc4H78u5X+RxNjLn5ZufZEoLxwBKCiXpX791Mk/Jz5F/ckyPN2/gdTmypPENYvBhKS30NxUQ/\nSwgm6g3dtR4I6DVjvOvaFbKy4NQpWLHC72hMlLOEYKLesB3rAPjYqovqx6qNjEeWEEx0Kyqib9E2\nzjZKYVX6FX5HE5tuusl5toRgamEJwUS3+fMBWJnel5LGTXwOJvZk5MzhsoVnKE1qRMXqXK76wQy/\nQzJRzBKCiW7z5gGwuOfVPgcSu86kNGVN2mUkoVy7a4Pf4ZgoZgnBRK/y8vNnCEssITRI5XAfw3au\n8zkSE80sIZjolZcHR46wq20ndrTr4nc0MW1JjwEAjNi+xoaxMNXylBBEZLSIbBGRAhHJqWK9iMiz\n7vqNIjLAXd5bRNYHPIrd22siIk+IyN6AdWND+9ZMzHOri5b0uNqGq2ig/I69KGrZni4nDsEmu9ey\nqVqtCUFEkoHngDFAFjBRRLKCio0BMt3HJOB5AFXdoqr9VLUfcDVwGng7YLtnKte7t9o05oL33gOs\n/SAkRFjcwz2Oc+b4G4uJWl7OEAYBBaq6XVVLgRlA8E1axwN/VsdKoK2IdA4qMxLYpqq7Ghy1iX+H\nD8Pq1ZCScuHuX6ZBPug10JmYa7+9TNW8JISuwJ6A+UJ3WV3LTABeD1r2oFvF9IqItKtq5yIySUTy\nRCTv4MGDHsI1ceH995267mHDOJ3SzO9o4sLHGf0oTWoEy5c7Q4kbEyQijcoikgKMA94MWPw80BPo\nB+wDnq5qW1V9QVWzVTU7NTU17LGaKOG2HzBmjL9xxJGTTZqTm54FFRVOwjUmiJeEsBdID5hPc5fV\npcwYYK2qFlUuUNUiVS1X1QrgRZyqKWOcLyy3uymjR/sbS5z5sGe2M2HtCKYKXhJCLpApIj3cX/oT\ngFlBZWYB33J7Gw0GjqvqvoD1EwmqLgpqY7gD+KTO0Zv4tGYNFBVBerozMJsJmQ8r2xHmzXOu8zAm\nQKPaCqhqmYhMAeYDycArqpovIpPd9dOAucBYoACnJ9G/VW4vIi2Am4DvBb30L0WkH6DAzirWm0T1\nzjvO87hx1t00xLa1T4MePWDHDsjNhcGD/Q7JRJFaEwKA2yV0btCyaQHTCjxQzbangIuqWH5PnSI1\niWOWewI6Prgzm2kwEf7U7nLu27GD3z3yO54ZdpidU2/xOyoTJexKZRNdduxwLpxq1Qquv97vaOLS\nYrcd4cZtuT5HYqKNJQQTXWbPdp7HjIGUFH9jiVMruvXldOMm9C3aRqfiQ36HY6KIJQQTXQLbD0xY\nlDRuwkfu2EY3b7W7qJkLLCGY6HH0KCxZAsnJdv1BmM3PHALAzVtX+hyJiSaWEEz0qOwKOXw4tG/v\ndzRxbdElgziXlMzg3ZvsqmVzniUEEz2suihiipu2ZGV6XxppBbz7rt/hmChhCcFEh9LS86Obcttt\n/saSIN6/1L0G4e23ay5oEoYlBBMdFi2C4mK4/HLo1cvvaBLC+5luQpg/H06f9jcYExU8XZhmTNi9\n6Yx7+Ezbq/hdjo2zEwlFrTqwvvOl9Nv3uTPY3e23+x2S8ZmdIRj/nTsHM2cCMKfPUJ+DSSzzL3V6\nG1m1kQFLCCYafPCB0+U0K4uCDt38jiahVHY/ZfZsJzGbhGYJwfjPrS7i61/3N44EtP2iNKfd5uhR\nWLjQ73CMzywhGH+dO3ehuuLOO/2NJVFNmOA8vx58Q0OTaCwhGH99+KFzYVSfPs4vVRN5d9/tPM+c\nCWfO+BuL8ZUlBOOvwOoiu/eBPzIz4eqr4cSJC9eCmIRkCcH4p7QU3nrLmbb2A99k5Mzhv1pfBcC7\nP3mGDOv2m7A8JQQRGS0iW0SkQERyqlgvIvKsu36jiAwIWLdTRDaJyHoRyQtY3l5EFojIVve5XWje\nkokZ773nVBf17es8jG/m9BkGwKhtq2lRYhepJapaE4KIJAPPAWOALGCiiATf6HYMkOk+JgHPB62/\nQVX7qWp2wLIcYJGqZgKL3HmTSP76V+f5m9/0Nw7DvtaprE7LomlZKaMKVvkdjvGJlzOEQUCBqm5X\n1VJgBhB8b8PxwJ/VsRJoKyKda3nd8cB0d3o6YJdJJpJjx5y+7yLwr//qdzQGmH3ZcADGf7rE50iM\nX7wkhK7AnoD5QneZ1zIKLBSRNSIyKaBMR1Xd507vBzpWtXMRmSQieSKSd/DgQQ/hmpjwz39CSQkf\nd+tLxu83kJEzx+qufTa391DOJSUzfMda2L/f73CMDyLRqDxUVfvhVCs9ICLDgwuoquIkjq9Q1RdU\nNVtVs1NTU8McqomYv/wFgJlZN/gciKl0uEVbPuw10BkS2/18TGLxkhD2AukB82nuMk9lVLXy+QDw\nNk4VFEBRZbWS+3ygrsGbGLV7NyxZwtlGKbzX+zq/ozEB3uw7ypl49VXQKn+jmTjmJSHkApki0kNE\nUoAJwKygMrOAb7m9jQYDx1V1n4i0EJFWACLSArgZ+CRgm3vd6XuBdxr4Xkys+NOfAFhwyTWcbNLc\n31jMl3zYM5uDzdvC5s2werXf4ZgIqzUhqGoZMAWYD2wG3lDVfBGZLCKT3WJzge1AAfAi8H13eUdg\nmYhsAFYDc1R1nrtuKnCTiGwFRrnzJo5l5Myh549nUfj07wGYcdXXfI7IBCtLbsTMy0c4M6++6mss\nJvI83Q9BVefifOkHLpsWMK3AA1Vstx24qprXPAyMrEuwJvYN27metOKD7G7TkeXdr/Q7HFOFN/uO\n4ru5M2HGDHjmGWjWzO+QTITYlcomoiZsmA84Zwcq9ucXjT5PzYCBA+H48QtXkpuEYP+RJmI6nDrK\nqIJVlEnShcZLE53uv995njat5nImrlhCMBFz56ZFNK4o54NLBnGwZXu/wzE1+cY3oFUrWLYMNm70\nOxoTIZYQTGSUl/ON9c5Imq9bY3L0a9UK7nU7AT73nL+xmIixhGAiY/Zs0o8XsbNtZ5b0GFB7eeO/\n77udBf/6V2eoERP3LCGYyPjtbwGYfvWtVCQl+xyM8eSyy+DGG+H0aZg+vfbyJuZZQjDht349LFnC\niZRmvNn3Jr+jMXXxgNub/LnnoKLC31hM2Hm6DsGYBnn2WQD+0XeUXZkcIyoHGkyuaMSS1qmkbd0K\ns2bB7TYocTyzMwQTXkVF8Le/gQh/uvo2v6MxdVSelMzLA90k8NRTNr5RnLOEYMLrmWegpATGjWNX\nuy5+R2PqYcaVX+No01awciUsXep3OCaMLCGY8Dly5EKXxZ/8xN9YTL2dSWnKnwfc6sw89ZS/wZiw\nsoRgwiIjZw6/HTcFTp7ko4z+ZLxlNzeKZX+6+lZnTKO5c2HTJr/DMWFiCcGERcuS09y3ZjYAv7/2\nbp+jMQ11tHkb+M53nJmf/czfYEzYWEIwYXHPujm0PXuS1WlZrE6/wu9wTCjk5EDTps7tT9eu9Tsa\nEwaWEEzoHT3K5JX/AOB/r53gczAmVDKeXccLfccA8MHXv+dzNCYcLCGY0Js6lTYlp/i4+5Uszejv\ndzQmhKYjOJONAAAOQUlEQVQNvpOTKc24cXseLF/udzgmxDwlBBEZLSJbRKRARHKqWC8i8qy7fqOI\nDHCXp4vIhyLyqYjki8gPArZ5QkT2ish69zE2dG/LRFpGzhwycuYw+PvTOfsbZ5iKp66/D0T8DcyE\n1JHmbXjl6nHOzOOP23UJcabWhCAiycBzwBggC5goIllBxcYAme5jEvC8u7wMeERVs4DBwANB2z6j\nqv3cx5fuyGZi0w+XvUbTslLe7T2UjZ0v9TscEwYvDbqDI81aw0cfwT/+4Xc4JoS8nCEMAgpUdbuq\nlgIzgPFBZcYDf1bHSqCtiHRW1X2quhZAVU/g3JO5awjjN1Hkqi+28PVNCylNasSvh9/jdzgmTIqb\ntrzw+T76qDP4nYkLXhJCV2BPwHwhX/1Sr7WMiGQA/YFVAYsfdKuYXhGRdh5jNlEoqaKcny+YRhLK\nywNvZ2d7y/vxbMaVN0O/frB7N/zqV36HY0IkIo3KItIS+CfwsKoWu4ufB3oC/YB9wNPVbDtJRPJE\nJO/gQbu4KVrdvXEBV+3fyhetOvC/dt1B3KtISj4/aCFTp8LOnb7GY0LDS0LYC6QHzKe5yzyVEZHG\nOMngNVU9f8duVS1S1XJVrQBexKma+gpVfUFVs1U1OzU11UO4JuL27eOxJX8C4Bc3fJvTKc38jcdE\nRMacYt657Ho4e5alI+4g47F3/Q7JNJCXhJALZIpIDxFJASYAs4LKzAK+5fY2GgwcV9V9IiLAy8Bm\nVf1N4AYi0jlg9g7gk3q/C+MfVfje92h79iSLe1zNnD5D/Y7IRNDPR36XI81aM2zXeu7auMDvcEwD\n1ZoQVLUMmALMx2kUfkNV80VksohMdovNBbYDBTi/9t1773EdcA9wYxXdS38pIptEZCNwA/DDkL0r\nEzmvvQazZ1PcpAU5ox+0bqYJ5nCLtjwxyrlI7T8+eAkKC32OyDSEaAz1I87Ozta8vDy/wzCVdu6E\n/v3h2DF+NOYHvHml3Q0tIany4lv/xU0Fq+D662HRIki226RGExFZo6rZtZWzK5VN/ZSWwt13Ozdf\nHzeON/uO8jsi4xcRckY/yIEW7WDJEnjySb8jMvVkCcHUT04OrF4N3brBq69aVVGCO9yiLQ/f+ojz\nd/Dzn8MHH/gdkqkHSwimzn546yPwzDOcS0rmjuEPkfHLFX6HZKLA8ox+8NOfOh0Nvv51KCjwOyRT\nR5YQTN0sXcrUeU7/85+P/C7ruvbxOSATVZ54Am691blb3m23OVWKJmZYQjDeffYZ3HEHTcrLePXq\n2/hL5W0VjamUnOz0PLviivN/L5w543dUxiNLCMabggK48UY4fJhFvQby5I3f8TsiE61at4bZs6Fz\nZ1i8GO680+mEYKKedTs1tdu+HW64wRm3ZsQI+gyYwtnGTf2OykS5Sw7tZuHM/wuHD8Ptt8Prrzt3\nXDMRZ91OTWisWwfXXuskgyFDYPZsSwbGk4IO3eD996FNG5g5E265BU6c8DssUwNLCKZ68+ZxYvB1\nUFTEx92v5IprfkjGfy3xOyoTSwYMcO6b0KmT0xV1+HAbCC+KWUIwX1VRAT/7GYwdS6vSM8zuM4x/\nu/NnnGzS3O/ITCy68kr4+GPo1QvWr4fsbOdqZhN1LCGYL9uzB8aMcboPAk8P/QYPjfsRpY0a+xuX\niW09ezoXMo4e7bQp3Hwz/OQnUFLid2QmgDUqG0dFBbz4IvzoR049b/v28Le/kfFhmd+RmTiSVFHO\n9ia58ItfOBewXX45vPwyXHON36HFNWtUNt598AEMHAiTJ8OJE8y7dAgDJ/7WkoEJuYqkZGeso6VL\nITMT8vNh8GD45jedjgvGV5YQEpUq99z9JEsz+sPIkbB2LftbtueBcY8x+fafcLBle78jNPHsuuuc\n9oScHGjSxLmY7dJL4Xvfg61b/Y4uYVmVUaI5fBhmzIA//hE2bQLgREozpl1zJy8PHG9dSk3EpR0v\n4sdLpnPr5qUkoVQgJI27De6/H8aOhcbWftVQXquMLCEkgi++gPfeg1mznOdz5wA40KIdr2aP47V+\nYyhu2tLnIE2i63m4kEmr3+KO/A9oUu5UVx5q3ob3M4ew6JKBvDz9MWjRwucoY1NIE4KIjAZ+ByQD\nL6nq1KD14q4fC5wG7lPVtTVtKyLtgb8DGcBO4C5VPVpTHJYQPKiogIICfvjvf6T/F1sYWJjPZQd3\nnl9dLkkszejPW1fcwLxLr7PeQybqpJ48yu35H3LnJwvpfSigXaFJE+fiyCFDnIslBw6Eiy+2odc9\nCFlCEJFk4HPgJqAQ5x7LE1X104AyY4EHcRLCNcDvVPWamrYVkV8CR1R1qojkAO1U9bGaYrGEgNMz\no7jY+dW/dy/s3ctvXlpAr8OF9DpSSM8jhTQ/9+WufKcbN+Hj7lexpGc28zMHW/uAiQ2qXH5gOzcW\nrGbktlz67d/q/P0HuugiyMpyHpmZTPnoAPtadWB/qw4caNmOc8mN2Tn1Fn/ijyKhTAhDgCdU9Wvu\n/OMAqvo/AWX+CCxW1dfd+S3ACJxf/1VuW1lGVfeJSGd3+941xVLvhPDJJ86Ii6oX/qAqpwMfVS0P\nR9nSUudRUuI8KqcDl506BcePX3gcO+Y8Hz3qrKvB/pbtWd+lN+u69GZ9596s69LHzgRMzNv5yCBY\nsQKWL3ee1693fhzVoLhJC1p3ToV27ZxH27bOUBrNmjnjKgU+V04nJ194NGr05fngR1JAv5zKM5XA\nM5ZQLKuc7tQJ0tI8Hq0v85oQGnl4ra7AnoD5QpyzgNrKdK1l246qus+d3g909BBL/dx1F2zeHLaX\nj7jmzaFrV+fRpQvTtpWwrX0aBRels+2iNGsPMHEp4+nVODXPw2DIMBisdDx5mMxDe8g8vJv0Y0V0\nOnGIzicO0+nEIS4+dZTWJadg56n4GC7jscdg6tTayzWAl4QQdqqqIlLlqYqITAImubMn3TOL+ugA\nHKrntuFU97hOn3a65oW3e178HK/IsLjqJiRx7QJWNzyWQNF6vOCppzrw1FP1ja27l0JeEsJeID1g\nPs1d5qVM4xq2LRKRzgFVRgeq2rmqvgC84CHOGolInpdTpkizuOrG4qobi6tuojUuiExsXi5MywUy\nRaSHiKQAE4BZQWVmAd8Sx2DguFsdVNO2s4B73el7gXca+F6MMcY0QK1nCKpaJiJTgPk4FXivqGq+\niEx2108D5uL0MCrA6Xb6bzVt6770VOANEfk2zpnfXSF9Z8YYY+rEUxuCqs7F+dIPXDYtYFqBB7xu\n6y4/DIysS7AN1OBqpzCxuOrG4qobi6tuojUuiEBsMXWlsjHGmPCxwe2MMcYAcZYQROTrIpIvIhUi\nkh207nERKRCRLSLytWq2by8iC0Rkq/vcLgwx/l1E1ruPnSKyvppyO0Vkk1su7Jdni8gTIrI3ILax\n1ZQb7R7DAvcK83DH9SsR+UxENorI2yLStppyETletb1/t2PFs+76jSIyIFyxBOwzXUQ+FJFP3b//\nH1RRZoSIHA/4fP8z3HG5+63xc/HpePUOOA7rRaRYRB4OKhOR4yUir4jIARH5JGCZp++hsPwvqmrc\nPIDLgN7AYiA7YHkWsAFoAvQAtgHJVWz/SyDHnc4BngpzvE8D/1nNup1AhwgeuyeAR2spk+weu55A\nintMs8Ic181AI3f6qeo+k0gcLy/vH6dzxXuAAIOBVRH47DoDA9zpVjjDxQTHNQJ4N1J/T14/Fz+O\nVxWf6X6gux/HCxgODAA+CVhW6/dQuP4X4+oMQVU3q2pVF66NB2aoaomq7sDpDTWomnLT3enpwO3h\nifT8gIB3Aa+Hax9hMAgoUNXtqloKzMA5ZmGjqu+rauWdelbiXMviFy/vfzzwZ3WsBNq619mEjaru\nU3cwSVU9AWzGGSUgFkT8eAUZCWxT1V0R3Od5qvoRcCRosZfvobD8L8ZVQqhBdUNrBIvccBowDChS\n1eouN1ZgoYisEedq7Uh40D1tf6Wa01SvxzFc7sf5NVmVSBwvL+/f12MkIhlAf2BVFauvdT/f90Tk\n8giFVNvn4vff1ASq/1Hmx/ECb99DYTluUTF0RV2IyEKgUxWrfqqqIbu4TbX64TRq4zHGidR8djBU\nVfeKyMXAAhH5zP01UW81xQU8DzyJ8w/8JE511v0N2V8o4qo8XiLyU6AMeK2alwn58Yo1ItIS+Cfw\nsKoGj/q2Fuimqifd9qGZQGYEworaz0Wci2XHAY9Xsdqv4/UlDfkeqo+YSwiqOqoem3kZfgM8DqfR\n0BhFpBHwL8DVNbzGXvf5gIi8jXOK2KB/JK/HTkReBN6tYpXX4xjSuETkPuBWYKS6FahVvEbIj1cV\nGjKMS1iJSGOcZPCaqr4VvD4wQajqXBH5g4h0UNWwjtvj4XPx5Xi5xgBrVbUoeIVfx8vl5XsoLMct\nUaqMZgETRKSJiPTAyfRVjYkVqeE0RgGfqWphVStFpIWItKqcxmlY/aSqsqESVG97RzX78zKMSajj\nGg38GBinqqerKROp49WQYVzCxm2PehnYrKq/qaZMJ7ccIjII53//cJjj8vK5RPx4Baj2LN2P4xXA\ny/dQeP4Xw92KHskHzhdZIVACFAHzA9b9FKdVfgswJmD5S7g9koCLgEXAVmAh0D5Mcf4JmBy0rAsw\n153uidNrYAOQj1N1Eu5j9xdgE7DR/cPqHByXOz8WpxfLtgjFVYBTV7refUzz83hV9f6ByZWfJ05v\nmefc9ZsI6O0WxpiG4lT1bQw4TmOD4priHpsNOI3z10Ygrio/F7+Pl7vfFjhf8G0ClkX8eOEkpH3A\nOfe769vVfQ9F4n/RrlQ2xhgDJE6VkTHGmFpYQjDGGANYQjDGGOOyhGCMMQawhGCMMcZlCcEYYwxg\nCcEYY4zLEoIxxhgA/j8EKO5h/BXUYgAAAABJRU5ErkJggg==\n", "text/plain": [ "" ] }, "metadata": {}, "output_type": "display_data" } ], "source": [ "mynorm = st.norm(0,2) # normal RV with mean 0 and variance 4, N(0,4)\n", "X = st.norm.rvs(0,2,size=10000) # Getting 10000 samples from N(0,4)\n", "plt.hist(X,50,normed=True)\n", "plt.plot(r,mynorm.pdf(r),'r',linewidth=2.0)" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "#### _$\\color{blue}{Exercise\\ 3}$_: \n", "\n", "In the review test, we saw that if $$X\\sim N(1,1)$$ and $$Y\\sim N(0,3^2),$$ and $X$ and $Y$ are jointly Gaussian with correlation coefficient $\\rho$, then $$Z=2X+3Y-2\\sim N(0,85+36\\rho).$$ \n", "\n", "Suppose that $\\rho=1/2$. Sample from the joint distribution for $X$, $Y$. To do this you need to use [scipy.stats.multivariate_normal](https://docs.scipy.org/doc/scipy-0.14.0/reference/generated/scipy.stats.multivariate_normal.html). For each sample compute $Z$. Plot the histogram for these samples and verify that they match $N(0,103)$. " ] }, { "cell_type": "code", "execution_count": 8, "metadata": {}, "outputs": [ { "data": { "text/plain": [ "[]" ] }, "execution_count": 8, "metadata": {}, "output_type": "execute_result" }, { "data": { "image/png": "iVBORw0KGgoAAAANSUhEUgAAAYQAAAD8CAYAAAB3u9PLAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nAAAAAElFTkSuQmCC\n", "text/plain": [ "" ] }, "metadata": {}, "output_type": "display_data" } ], "source": [ "# We need to first create a multivariate normal random variable.\n", "# For that we need to have its mean and covariance matrix\n", "m = [1,0]\n", "K = [[1,1*3*1/2],[1*3*1/2,3**2]]\n", "XYRV = st.multivariate_normal(m,K)\n", "# sample from distribution\n", "XY = XYRV.rvs(size=10000)\n", "\n", "Z = [2*x+3*y-2 for [x,y] in XY]\n", "ZRV = st.norm(0,np.sqrt(103))\n", "r = np.linspace(-40,40)\n", "plt.hist(Z,50,normed=True)\n", "plt.plot(r,ZRV.pdf(r),'r',linewidth=2.0)" ] }, { "cell_type": "code", "execution_count": null, "metadata": { "collapsed": true }, "outputs": [], "source": [] } ], "metadata": { "anaconda-cloud": {}, "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": "3.6.0" } }, "nbformat": 4, "nbformat_minor": 1 }