nvidia-AI.summerschool17-no.../teaching-ncso/97-sequence-modeling.ipynb

759 lines
440 KiB
Text
Raw Permalink Normal View History

2017-09-22 17:38:09 +00:00
{
"metadata": {
"name": "97-sequence-modeling"
},
"nbformat": 3,
"nbformat_minor": 0,
"worksheets": [
{
"cells": [
{
"cell_type": "heading",
"level": 1,
"metadata": {},
"source": [
"Time Series and Forecasting"
]
},
{
"cell_type": "heading",
"level": 2,
"metadata": {},
"source": [
"Airline Passengers"
]
},
{
"cell_type": "code",
"collapsed": false,
"input": [
"passengers = 1.0*array([112, 118, 132, 129, 121, 135, 148, 148, 136, 119, 104, 118, 115, 126, 141, 135, 125, 149, 170, 170, 158, 133, 114, 140, 145, 150, 178, 163, 172, 178, 199, 199, 184, 162, 146, 166, 171, 180, 193, 181, 183, 218, 230, 242, 209, 191, 172, 194, 196, 196, 236, 235, 229, 243, 264, 272, 237, 211, 180, 201, 204, 188, 235, 227, 234, 264, 302, 293, 259, 229, 203, 229, 242, 233, 267, 269, 270, 315, 364, 347, 312, 274, 237, 278, 284, 277, 317, 313, 318, 374, 413, 405, 355, 306, 271, 306, 315, 301, 356, 348, 355, 422, 465, 467, 404, 347, 305, 336, 340, 318, 362, 348, 363, 435, 491, 505, 404, 359, 310, 337, 360, 342, 406, 396, 420, 472, 548, 559, 463, 407, 362, 405, 417, 391, 419, 461, 472, 535, 622, 606, 508, 461, 390, 432])\n",
"dates = linspace(1949,1961,len(passengers)+1)[:-1]"
],
"language": "python",
"metadata": {},
"outputs": [],
"prompt_number": 78
},
{
"cell_type": "code",
"collapsed": false,
"input": [
"plot(dates,passengers)"
],
"language": "python",
"metadata": {},
"outputs": [
{
"output_type": "pyout",
"prompt_number": 15,
"text": [
"[<matplotlib.lines.Line2D at 0x3d95dd0>]"
]
},
{
"output_type": "display_data",
"png": "iVBORw0KGgoAAAANSUhEUgAAAX4AAAD9CAYAAAC7iRw+AAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nfiLyC39l/K4Cv9EtnXoCf9Ds
}
],
"prompt_number": 15
},
{
"cell_type": "heading",
"level": 2,
"metadata": {},
"source": [
"LR04 Stack"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"Please cite: Lisiecki, L. E., and M. E. Raymo (2005), A Pliocene-\n",
"Pleistocene stack of 57 globally distributed benthic d18O records, \n",
"Paleoceanography,20, PA1003, doi:10.1029/2004PA001071."
]
},
{
"cell_type": "code",
"collapsed": false,
"input": [
"lr04 = genfromtxt(\"lisiecki-lr04stack.txt\",comments=\"#\")\n",
"ix = argsort(-lr04[:,0])\n",
"lr04time = -lr04[ix,0]\n",
"lr04d180 = -lr04[ix,1]\n",
"plot(lr04time,lr04d180)"
],
"language": "python",
"metadata": {},
"outputs": [
{
"output_type": "pyout",
"prompt_number": 40,
"text": [
"[<matplotlib.lines.Line2D at 0x5af2c90>]"
]
},
{
"output_type": "display_data",
"png": "iVBORw0KGgoAAAANSUhEUgAAAXoAAAD9CAYAAACyYrxEAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n+FD53/3O2+aBA9UTCN0/PV0J
}
],
"prompt_number": 40
},
{
"cell_type": "code",
"collapsed": false,
"input": [
"plot(lr04time[-500:],lr04d180[-500:])"
],
"language": "python",
"metadata": {},
"outputs": [
{
"output_type": "pyout",
"prompt_number": 42,
"text": [
"[<matplotlib.lines.Line2D at 0x5b5be10>]"
]
},
{
"output_type": "display_data",
"png": "iVBORw0KGgoAAAANSUhEUgAAAXoAAAD9CAYAAACyYrxEAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz\nAAALEgAACxIB0t1+/AAAIABJREFUeJztnXt8FOX1/z8bkgC5AZFwSRYMEJKAkIsEqLVoKgREBIu2\nlWJra9W2tLZVtLWWqtB+g5dK+63Qfmu1YrW1P3tR04ogwRrbaiEVAQEpYAmQOyEk5H5lfn8cnszs\n7uzu7Ozs7szseb9eee1tdvfJ7MxnznPOec5xSJIkgWEYhrEtMZEeAMMwDBNaWOgZhmFsDgs9wzCM\nzWGhZxiGsTks9AzDMDaHhZ5hGMbm6Bb6Bx98EPn5+SgoKMDChQtRXV2tut2OHTuQm5uL6dOn47HH\nHtM9UIZhGEYfDr159O3t7UhOTgYAbN68GQcOHMAzzzzjss3g4CBycnKwa9cuZGRkYO7cufj973+P\nGTNmBD9yhmEYRhO6LXoh8gDQ0dGBsWPHemxTWVmJrKwsZGZmIi4uDqtWrUJZWZner2QYhmF0EBvM\nm9etW4cXXngBCQkJ2L17t8frtbW1mDRp0tBjp9OJPXv2BPOVDMMwTID4FPqSkhI0NDR4PL9x40Ys\nX74cpaWlKC0txaOPPop77rkHW7duddnO4XBoHkgg2zIMwzAy/jzwPoW+vLxc05esXr0a1113ncfz\nGRkZLkHa6upqOJ1Or5/DZXeI9evXY/369ZEehingfSHD+0KG94WMFiNZt4/++PHjQ/fLyspQWFjo\nsU1RURGOHz+OkydPoq+vDy+99BJWrFih9ysZhmEYHej20T/wwAM4evQohg0bhmnTpuH//u//AAB1\ndXW48847sW3bNsTGxmLLli1YsmQJBgcHcfvtt3PGDcMwTJjRnV5pNA6Hg103F6moqEBxcXGkh2EK\neF/I8L6Q4X0ho0U7WegZhmEsjBbt5BIIDMMwNoeFnmEYxuaw0JuAwcFIj4Bhwkd5OXDPPZEeRXTB\nQh9hjh0Drrgi0qNgmPBx6hSwf3+kRxFdsNBHmKoq+mOYaKGzE1BZcM+EEBb6CNPQADQ3s/uGiQ7e\nfx+orWWhDzdBFTVj9PHBB0BsLDBzJh3wkgT86lfArbcCiYmRHh3DhIYLF4A5c4C4OKC/H+jpAUaM\niPSoogO26CPAb35Df4Bs2Xz968DPfha5MTFMqHnnHbodNoxuGxsjN5Zog4U+ArS1UTBq/XrXKez2\n7REbEsOEHFEeq6eHbtl9Ez5Y6CNAezuwaxfwxBPkrxwzBpg1C9i3j15jGDvS3e36mIVePzt2AJ/+\ntPbtWegjQFsb+Ss7O4H33gMKC4G8PKCoSJ7eWpFt24ApUyI9CsasdHcDMRcVZ/x4FvpgKCsD/vxn\n7duz0EeAtja6dTiASy+lK/OVV9KflRtwHTwInDwpP+bZCaOkuxuYMIHuT5vGQh8MonOr1vJgLPQR\nQAjg6tXAd74DrFlDwdjx4ynV0qokJdGt6CqZkgKcPRu58TDmQin0WVks9MEg0rHr6rRtz+mVEaCt\nDVi1CnjySSA1VX5+9GigtTVy4wqWlha6veIK4Kmn6H5fX+TGw5gLd4t+377IjsfKiHPtYhsQv7DQ\nR4C2NmDLFleRB+wh9MXFwMKFwGOP0XO9vREdEmMi3C16zjLTT0sLsHw5cPq0tu3ZdRNmJIlcN8nJ\nnq9ZXejPnaNFX5/9rFzWgYWeEXR1sY/eKM6dA77xDeD557VtzxZ9mOntpcyD+HjP16wu9C0tlCo6\nYYIcJBI50wzj7rqprwc6OuTYDuOfI0coBibONa2wRR9m2tooSKmG1YX+3Dk6+JKTgZEj6Tm26BmB\nEPrEROCSSygZYd26SI/KWlRUAJs3k9C7u359wUIfZtra1N02gPWFXhx8DodsubHQM4LubjrGq6ro\nGPnudykfnNFOTQ1w+DCVj2CL3sRUV8s5sO4kJZEfc2AgvGMygk9/mg5AcfAJoWfXDSPo7gYSEoC0\nNHqck0PHh3LtBeOb2lrKZOvooIumVljow8yLLwI33qj+WkwMWftiQZWVOHWKbsVJzBY94053t+zS\nA8iqv+IK4N//jtyYrEZtLTB9OqVmi+JwWmChDzNvvAHcdJP3163qvomNpfINw4fT42XLgHHjWOgZ\nGXehB8hXf/58ZMZjRWpqgFdeAb75zcDex1k3YWRwkFLKLr3U+zajRllT6Jub6aQV3H478Le/seuG\nkVETeqvOYCNFbS3gdAb+Prbow0hjI4lhXJz3bZKSqNiZ1XAXeoCse7boGYGa0KeksNBrpaGBtMNb\n1p4vWOjDSE0NkJHhe5ukJAq0WInBQZp+u2cBjBjBFj0j09WlLvRc/E4b+/dTpVuHI/D3stCHkdpa\newp9SwudsO7BIbboGYEksUUfLPv3AwUF+t7LQh9Gamr8+9esKPTNzeopoyz0jKC/nyxRd7cl++i1\nc+AAkJ+v770s9GGkqgqYNMn3NlYU+txcWh/gDrtuGEFfn5yRpYRdN/4ZGKCL5AcfADNm6PsMFvow\n8s9/Ah//uO9trCb0ogzx9dd7vsYWPSPo71ev78SuG/+IC+GHHwLZ2fo+g4U+TFRWUkGi+fN9b2c1\noW9ro7IHf/yj52ss9Iygr08924xdN/4R+2fCBH0ZNwALfdi45hrgS19Sn74qsZrQnz9Puf9qsOuG\nEfT1ebfo2XXjGyH0OTn6P4OFPgxIEvnZHn/c/7Z2Enq26BkBu270095OpUXuuUf/Z7DQh4HOTqpj\n455apoYVhd7bdNLuFv1vf0s9fxn/+HPdaG1yHY20tVH+/A036P8MFvowoLZq1BtWE/q2tui16H/1\nK+CJJ9j1oAVvFn18PP1ZcTV4uPDVw0IrLPRhwM5CH82uG1GMq7Y2suOwAt589ACdG2fPhnc8VsJX\nDwut6Bb6Bx98EPn5+SgoKMDChQtRrZZIDSAzMxN5eXkoLCzEvHnzAvqO224DDh7UO0Lz4G1BkRrJ\nyST0Im/W7PgS+oQEe1tqzc3UEm/GDOAPf4j0aMyNN9cNQOdGc3N4x2Ml2tsjaNF/97vfxYEDB7B/\n/3586lOfwoYNG1S3czgcqKiowL59+1BZWRnQdxw4YB+h12rRp6TI1Su1dniPJL5cN04nrQa2K83N\n8krFQ4cCe+9vf6uvZolV8ea6AejcYKH3TkRdN8mKuURHRwfG+jBZJZ2Rlo4OewjF2bPahV4pjsFO\n18KBL4t+8mRaMXvhQnjHFA66uuh2+nS6TU8P7P3R5u7x57phofeOEa6boOrRr1u3Di+88AISEhKw\ne/du1W0cDgcWLVqEYcOG4atf/SruvPNOr5+3fv36ofvFxcXo6ChWXVpvNQKx6EeOpAVIDQ3yqlMz\nc/48MHWq+msjR1JFy/p6/8XcrIb4TcX/FWgsYvx4ulUr9GVHfLluWOh9097uuiK2oqICFRUVAX2G\nT6EvKSlBQ0ODx/MbN27E8uXLUVpaitLSUjz66KO45557sHXrVo9t33nnHUycOBFNTU0oKSlBbm4u\nFixYoPp9SqEHyKK3g9A3NACXXaZ9+8xMeo8VUhM7OiiA7I3MTOoJalehX7wYyMoKPBYh+gJ/9BEw\ne7bx4zMb/lw3HIz1jrvrpri4GMXFxUOPvbnNlfgU+vLyck0DWb16Na677jrV1yZOnAgASEtLw8qV\nK1FZWelV6JVIkn2E/uRJ9Vow3sjMBHbvtobQ9/T4tkgzM4Gf/IRKP8TaqJ+ZEPqcHODWW2VXjla6\nu+n24EH/Qr9rF7B9O7Bxo/+V1WbFn+vmo4/COx4r0dzs2eshUHT76I8fPz50v6ysDIWFhR7bdHV1\nof1iknFnZyd27tyJ2RrNl+5uEnu7CH1mpvbtp0yhWyukJvb00MIob3zve8DLL9MMxU4o4y4JCYEL\nfU8P+fe//31Z9L3xwgt0sTxwQN9YzYC/rBu26L3T0ABctJd1o1voH3jgAcyePRsFBQWoqKjApk2b\nAAB1dXVYtmzZxQE2YMGCBSgo
}
],
"prompt_number": 42
},
{
"cell_type": "code",
"collapsed": false,
"input": [
"plot(lr04time)\n",
"len(lr04time)"
],
"language": "python",
"metadata": {},
"outputs": [
{
"output_type": "pyout",
"prompt_number": 63,
"text": [
"2115"
]
},
{
"output_type": "display_data",
"png": "iVBORw0KGgoAAAANSUhEUgAAAY0AAAD9CAYAAABA8iukAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nK5k1axYAhYWFnH/++dx0000D
}
],
"prompt_number": 63
},
{
"cell_type": "code",
"collapsed": false,
"input": [
"from scipy.interpolate import interp1d\n",
"f = interp1d(lr04time,lr04d180)\n",
"d180t = linspace(-5000,0,5000)\n",
"d180 = f(d180t)\n",
"plot(d180)"
],
"language": "python",
"metadata": {},
"outputs": [
{
"output_type": "pyout",
"prompt_number": 66,
"text": [
"[<matplotlib.lines.Line2D at 0x8b82d50>]"
]
},
{
"output_type": "display_data",
"png": "iVBORw0KGgoAAAANSUhEUgAAAYMAAAD9CAYAAABeOxsXAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nbJxwtrntNpqwtXs3MGAAjdd/
}
],
"prompt_number": 66
},
{
"cell_type": "heading",
"level": 2,
"metadata": {},
"source": [
"Stock Market"
]
},
{
"cell_type": "code",
"collapsed": false,
"input": [
"stocks = genfromtxt(\"stock-market.csv\",delimiter=\",\")\n",
"days = stocks[:,0]\n",
"usdeur = stocks[:,2]\n",
"djia = stocks[:,4]\n",
"aapl = stocks[:,6]\n",
"msft = stocks[:,8]\n",
"nasdaq = stocks[:,10]\n",
"plot(days,aapl/mean(aapl))\n",
"plot(days,djia/mean(djia))\n",
"plot(days,nasdaq/mean(nasdaq))"
],
"language": "python",
"metadata": {},
"outputs": [
{
"output_type": "pyout",
"prompt_number": 48,
"text": [
"[<matplotlib.lines.Line2D at 0x6931e50>]"
]
},
{
"output_type": "display_data",
"png": "iVBORw0KGgoAAAANSUhEUgAAAXsAAAD9CAYAAABdoNd6AAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nWFxDQwE3N83ty0PsGzdWi33O
}
],
"prompt_number": 48
},
{
"cell_type": "heading",
"level": 1,
"metadata": {},
"source": [
"Stationarity, Periodicity"
]
},
{
"cell_type": "code",
"collapsed": false,
"input": [
"plot(passengers)"
],
"language": "python",
"metadata": {},
"outputs": [
{
"output_type": "pyout",
"prompt_number": 79,
"text": [
"[<matplotlib.lines.Line2D at 0xa37f650>]"
]
},
{
"output_type": "display_data",
"png": "iVBORw0KGgoAAAANSUhEUgAAAXsAAAD9CAYAAABdoNd6AAAABHNCSVQICAgIfAhkiAAAAAlwSFlz\nAAALEgAACxIB0t1+/AAAIABJREFUeJzt3XtYlGX+P/D3IJit5xODMuiYnBw8YZ6yrTBEq02iNAoP\nsda2u/Ft2/rVlnldXavtJrjWflcrvrVlyVa7aFsr5CqLJGRlhoZmObagjspxVBAPqA0w9++PDw8z\nwByemXmGGZjP67q4gjncc0P1nns+z31QCSEEGGOM9WpBvu4AY4wx7+OwZ4yxAMBhzxhjAYDDnjHG\nAgCHPWOMBQAOe8YYCwBOw/6///0v4uPj278GDx6MjRs3oqGhAUlJSYiOjsb8+fPR2NjY/pzMzExE\nRUUhNjYWhYWFXv0FGGOMOadyZZ692WxGeHg4SktL8eqrr2LEiBF49tlnsW7dOpw/fx5ZWVnQ6/VY\nsmQJ9u/fj+rqasybNw/l5eUICuIPEYwx5isuJXBRUREiIyMRERGB/Px8pKenAwDS09Oxbds2AEBe\nXh7S0tIQEhICrVaLyMhIlJaWKt9zxhhjsrkU9rm5uUhLSwMAGI1GqNVqAIBarYbRaAQA1NTUQKPR\ntD9Ho9Ggurpaqf4yxhhzQ7DcB5pMJnzyySdYt25dl/tUKhVUKpXd53a+z9FjGWOM2efuDjeyR/Y7\nd+7EjTfeiJEjRwKg0XxdXR0AoLa2FqGhoQCA8PBwVFZWtj+vqqoK4eHhNjvs71+///3vfd4H7if3\ns6f2kfup/JcnZIf9P/7xj/YSDgAkJycjJycHAJCTk4OUlJT223Nzc2EymWAwGFBRUYGZM2d61EnG\nGGOekVXGaWpqQlFREd56663221auXInU1FRs2rQJWq0WW7duBQDodDqkpqZCp9MhODgY2dnZXLZh\njDEfkxX2/fv3x7lz5zrcNmzYMBQVFdl8/KpVq7Bq1SrPe+djCQkJvu6CLNxPZfWEfvaEPgLcT3/i\n0jx7xV5UpfK4/sQYY4HGk+zklU6MMRYAOOwZYywAcNgzxlgA4LBnjLEAwGHPGGMBgMOeMcYCAIc9\nY4wFAA57xhgLABz2jDEWADjsGWMsAHDYM8ZYAOCwZ4yxAMBhzxhjAYDDnjHGAgCHPWOMBQAOe8YY\nCwAc9owxFgA47BljLABw2DPGWADgsGeMsQDAYc8YYwGAw54xxgIAhz1jLKC8/DLwz3/6uhfdj8Oe\nMRZQ9u8HsrN93Yvux2HPGAsojY1ASQlQXe3rnnQvDnvGWEA5fx6YMAHYssXXPeleHPaMsYBy/jyQ\nkQH8/e++7kn3khX2jY2NWLx4MSZMmACdToevv/4aDQ0NSEpKQnR0NObPn4/Gxsb2x2dmZiIqKgqx\nsbEoLCz0WucZY8xVjY3AffdRGae83Ne96T6ywv63v/0t7rrrLhw9ehSHDx9GbGwssrKykJSUhPLy\nciQmJiIrKwsAoNfrsWXLFuj1ehQUFCAjIwNms9mrvwRjjMkhBIX98OHAokXAtm2+7lH3cRr2Fy5c\nwOeff46HH34YABAcHIzBgwcjPz8f6enpAID09HRsa/ur5eXlIS0tDSEhIdBqtYiMjERpaakXfwXG\nGJOnqQno25e+xo4FjEZf96j7OA17g8GAkSNHYsWKFZg2bRoeffRRNDU1wWg0Qq1WAwDUajWMbX+1\nmpoaaDSa9udrNBpUB9plb8aYXzp/Hhg6lL4fPBi4cMG3/elOwc4e0NLSgrKyMrz22muYMWMGnnzy\nyfaSjUSlUkGlUtltw9Z9q1evbv8+ISEBCQkJ8nvNGGNuaGwEhgyh7wcPBi5e9G1/nCkpKUFJSYki\nbTkNe41GA41GgxkzZgAAFi9ejMzMTISFhaGurg5hYWGora1FaGgoACA8PByVlZXtz6+qqkJ4eHiX\ndq3DnjHGuoP1yH7QIP8f2XceCK9Zs8bttpyWccLCwhAREYHytsvWRUVFiIuLw8KFC5GTkwMAyMnJ\nQUpKCgAgOTkZubm5MJlMMBgMqKiowMyZM93uIGOMKaXzyN7fw15JTkf2APDqq69i6dKlMJlMGD9+\nPN599120trYiNTUVmzZtglarxdatWwEAOp0Oqamp0Ol0CA4ORnZ2tsMSD2OMdZfONXt/L+MoSSWE\nEN3+oioVfPCyjLEAt2EDcPw4sHEjUFUFzJrVs7ZN8CQ7eQUtYyxg9LSavZI47BljAcO6Zj9gAHD1\nKtDa6ts+dRcOe8ZYwLAe2QcFAQMHBk7dnsOeMRYwGhstYQ8EVimHw54xFjDOn7eUcYDAmpHDYc8Y\nCxjWZRwgsObac9gzxgKG9QVagMs4jDHWK9ka2XMZhzHGepHmZuDaNZpyKeEyDmOM+dCBA0BNjbJt\nSiUc691bOOwZY8xHfvyRjg3My1O23c71eoBr9owx5jNvvw1UViofwp3r9QDX7BljzCeuXgXWrqXz\nYZUOe1sjey7jMMaYD7zxBjBzJjB3bveM7LmMwxhjPrB9O/Doo94Zcdsb2QdKGUfW4SWMMdYd6uuB\nsDDaibK7avaBMrLnsGeM+Y36emD4cODyZe+M7LmMwxhjfkAKe2+MuDtvggYEVhmHw54x5hekg0T6\n9/dezd7e1MtAOCWVw54x5hekUb1K5Z2wP3uW2rcWEkJfV64o+1r+iMOeMeYXpLAHqJZ+6RJgNivX\nvtEIqNVdbx80KDBKORz2jDG/YB32ffoAP/kJXahVir2wD5QZORz2jDG/YB32gLIh3NJCbXUu4yj9\nOv6Mw54x5he8GfZSvb5Pn673cRmHMca6kTfD3mgEQkNt38cje8YY60beDntb9XqlX8efcdgzxvyC\nr8KeyziMMdaNGhp4ZO9NHPaMMb9QXw8MG2b5mcNeWbLCXqvVYvLkyYiPj8fMmTMBAA0NDUhKSkJ0\ndDTmz5+PxsbG9sdnZmYiKioKsbGxKCws9E7PGWM+cfiwd1ac+qqMExpK9/d2ssJepVKhpKQEBw8e\nRGlpKQAgKysLSUlJKC8vR2JiIrKysgAAer0eW7ZsgV6vR0FBATIyMmBWchkcY8xnWlqAO+8EiouV\nb9ubYX/mjP2wHzMGOH1amdfxZ7LLOKLTTkH5+flIT08HAKSnp2Pbtm0AgLy8PKSlpSEkJARarRaR\nkZHtbxCMsZ5t+3agpkb5C5pmM21U5s0yjr2pl2PHAqdOKfM6/kzWfvYqlQrz5s1Dnz598Ktf/QqP\nPvoojEYj1G1vlWq1Gsa2z0E1NTWYPXt2+3M1Gg2qq6u7tLl69er27xMSEpCQkODBr8EY6w5vvkmj\nb2/sSDlgABBslUjdVcYZPZrub26mTdH8SUlJCUpKShRpS1bYf/nllxg1ahTOnj2LpKQkxMbGdrhf\npVJBpVLZfb6t+6zDnjHm/06eBPbvB9LSlB/Zdy7hAMqFvdkMnDtnf2QfEkKnY1VXA1qt56+npM4D\n4TVr1rjdlqwyzqhRowAAI0eOxL333ovS0lKo1WrU1dUBAGpraxHa9pcMDw9HZWVl+3OrqqoQHh7u\ndgcZY/7hrbeAZctohNyTwr6+nubSOxq1B0Ipx2nYX7lyBZcuXQIANDU1obCwEJMmTUJycjJycnIA\nADk5OUhJSQEAJCcnIzc3FyaTCQaDARUVFe0zeBhjPde2bRT23jjKz17YW03yc5ujEo4kEC7SOi3j\nGI1G3HvvvQCAlpYWLF26FPPnz8f06dORmpqKTZs2QavVYuvWrQAAnU6H1NRU6HQ6BAcHIzs722GJ\nhzHWM1RVAePHA0eO9KyRvZywD4SRvdOwHzduHA4dOtTl9mHDhqGoqMjmc1atWoVVq1Z53jvGmF9o\nagJMJjrD1RvbC9gKe+sDTII8WP4pN+wPHHD/NXoCXkHLGHOqthYYNYqODOyusA8OBq6/3vMDTBxN\nu5QEQhmHw54x5lRNDU1RBLqvZg8oU8pxtKBKEghlHA57xphTtbWWsB88uHtG9tJreRr2rlyg7bR2\ntFfhsGeMOVVTQ2UcwDtlnM47
}
],
"prompt_number": 79
},
{
"cell_type": "code",
"collapsed": false,
"input": [
"# moving average\n",
"ma = [mean(passengers[i-12:i]) for i in range(12,len(passengers))]\n",
"plot(ma)"
],
"language": "python",
"metadata": {},
"outputs": [
{
"output_type": "pyout",
"prompt_number": 95,
"text": [
"[<matplotlib.lines.Line2D at 0xbd25850>]"
]
},
{
"output_type": "display_data",
"png": "iVBORw0KGgoAAAANSUhEUgAAAXsAAAD9CAYAAABdoNd6AAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nNC3T3WmBlEiAWbcOhg2Dvn3N
}
],
"prompt_number": 95
},
{
"cell_type": "code",
"collapsed": false,
"input": [
"# exponential smoothing\n",
"es = []\n",
"y = 0\n",
"for i,x in enumerate(passengers):\n",
" y = 0.2*x+0.8*y\n",
" es.append(y)\n",
"plot(es)"
],
"language": "python",
"metadata": {},
"outputs": [
{
"output_type": "pyout",
"prompt_number": 98,
"text": [
"[<matplotlib.lines.Line2D at 0xc5f6350>]"
]
},
{
"output_type": "display_data",
"png": "iVBORw0KGgoAAAANSUhEUgAAAXsAAAD9CAYAAABdoNd6AAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nPsz+/fuZOHEidrud8vJyQkND
}
],
"prompt_number": 98
},
{
"cell_type": "code",
"collapsed": false,
"input": [
"# Gaussian smoothing\n",
"from scipy.ndimage import filters\n",
"trend = filters.gaussian_filter(passengers,12)\n",
"plot(trend)"
],
"language": "python",
"metadata": {},
"outputs": [
{
"output_type": "pyout",
"prompt_number": 99,
"text": [
"[<matplotlib.lines.Line2D at 0xc620350>]"
]
},
{
"output_type": "display_data",
"png": "iVBORw0KGgoAAAANSUhEUgAAAXsAAAD9CAYAAABdoNd6AAAABHNCSVQICAgIfAhkiAAAAAlwSFlz\nAAALEgAACxIB0t1+/AAAIABJREFUeJzt3Xt0VOW9//H30MRjEZbgJYkngyYlCXHCbRQDWm1HSaBF\niUA0miDkILSnePkVwXuLTeoxCQVrgdVYamMNYglUC4kcoOHiIHJpQMB6mmqiTTR35RIRAYeQ/ftj\ny5QokJBksieZz2utWQ4ze2Y+ieSbh+9+9vPYDMMwEBGRHq2X1QFERMT3VOxFRAKAir2ISABQsRcR\nCQAq9iIiAUDFXkQkALSp2EdERDB06FCcTifx8fEAHDx4kMTERGJiYhgzZgyNjY3e47Ozs4mOjiY2\nNpbi4mLfJBcRkTZrU7G32Wy43W727t1LSUkJADk5OSQmJlJWVsbo0aPJyckBoLS0lBUrVlBaWsr6\n9eu57777aG5u9t1XICIirWpzG+fr114VFRWRnp4OQHp6OqtXrwagsLCQ1NRUgoODiYiIICoqyvsL\nQkRErNHmkX1CQgIjRozghRdeAKChoYHQ0FAAQkNDaWhoAKC2tha73e59rd1up6amprNzi4jIeQhq\ny0Hbtm3jiiuu4NNPPyUxMZHY2NgWz9tsNmw221lf//XnznWsiIicXXtXuGnTyP6KK64A4PLLL2fi\nxImUlJQQGhpKfX09AHV1dYSEhAAQHh5OVVWV97XV1dWEh4efMbC/337xi19YnkE5lbO7ZlTOzr91\nRKvF/ujRo3z++ecAfPHFFxQXFzNkyBCSkpLIz88HID8/nwkTJgCQlJREQUEBHo+HiooKysvLvTN4\nRETEGq22cRoaGpg4cSIATU1NTJ48mTFjxjBixAhSUlLIy8sjIiKClStXAuBwOEhJScHhcBAUFERu\nbq7aNiIiFmu12EdGRrJv375vPH7JJZewcePGM77mySef5Mknn+x4Oou5XC6rI7SJcnau7pCzO2QE\n5fQnNqOjjaD2fKjN1uH+k4hIoOlI7dRyCSIiAUDFXkQkAKjYi4gEABV7EZEAoGIvIhIAVOxFRAKA\nir2ISABQsRcRCQAq9iIiAUDFXkQkAKjYi4gEABV7EZEAoGIvIhIAVOxFRM7C47E6Qedp0x60IiI9\n1fHj8PbbUFIC77wD5eXw0Udw8CAMGADvv291ws6h9exFJOB8+im89hq8/jps3QoxMTByJDidEB0N\nERFw2WXQuzf400Z7HamdKvYiEhBOnoR16+B3vzML/LhxMHEiJCZC//5Wp2ubjtROtXFEpEc7dgz+\n+EdYsMAcrc+cCStWwEUXWZ2sa7XpBO3JkydxOp2MHz8egIyMDOx2O06nE6fTybp167zHZmdnEx0d\nTWxsLMXFxb5JLSLSiqYm+P3vYeBA+OtfYdkysy8/bVrgFXpo48h+4cKFOBwOPv/8c8D8p8Ts2bOZ\nPXt2i+NKS0tZsWIFpaWl1NTUkJCQQFlZGb16adKPiHQNwzB78Y89BmFhUFQEI0ZYncp6rVbh6upq\n1q5dy4wZM7y9IsMwztg3KiwsJDU1leDgYCIiIoiKiqKkpKTzU4uInME//wk33wxPPgnPPgubN6vQ\nn9JqsX/ooYeYP39+i9G5zWZj8eLFDBs2jOnTp9PY2AhAbW0tdrvde5zdbqempsYHsUVE/s3jgV/+\nEr73PbjzTti3zzwB608zaax2zjbOmjVrCAkJwel04na7vY/PnDmTp556CoC5c+cyZ84c8vLyzvge\ntrN8tzMyMrz3XS4XLpfr/JKLiAB/+xvMmAFXXQV79phz43sKt9vdovZ2xDmnXj755JO8/PLLBAUF\ncfz4cQ4fPkxycjJLly71HlNZWcn48eN59913ycnJAeDxxx8H4Ac/+AGZmZmMHDmy5Ydq6qWIdNCJ\nE5CRAXl58JvfwF139fyRfJfMs9+yZQsLFizg9ddfp66ujiuuuAKA5557jl27dvGnP/2J0tJS0tLS\nKCkp8Z6g/eCDD74xulexF5GO+Ne/IC3NnB//0ksQGmp1oq7RJfPsDcPwFu1HH32Ud955B5vNRmRk\nJEuWLAHA4XCQkpKCw+EgKCiI3Nzcs7ZxRETaY9kyeOgh+PnP4cEHQZP92kZX0IpIt3DsGNx/P+zc\nCcuXw7BhVifqeh2pnfqdKCJ+71//ghtuMBct27UrMAt9R6nYi4hfW7MGrr8e7r0XXnklMK9+7Qxa\nG0dE/FJzszl3Pi8PVq0yR/bSfir2IuJ3vvgC0tOhthZ27w6c2Ta+pDaOiPiVqiq48Ubo0wfeeEOF\nvrOo2IuI39ixw9xEZPJkc1ni//gPqxP1HGrjiIhfePllmDPHLPK33mp1mp5HxV5ELNXcbK5SuXKl\n2baJi7M6Uc+kYi8ilvn8c7Nlc/iwubHIZZdZnajnUs9eRCxRUWFOpwwLg+JiFXpfU7EXkS735ptm\nof/xj2HJErjgAqsT9Xxq44hIl8rLM3v0L78MY8ZYnSZwqNiLSJdoaoJHHoG1a82R/aBBVicKLCr2\nIuJzhw7B3XebM2927jTXoZeupZ69iPjUe++ZF0o5HLBunQq9VVTsRcRn1q41NwF/4gl47jkIUi/B\nMvrWi0inMwxYsMDcG3b1aq1Y6Q9U7EWkUx07Zk6pLC01+/MDBlidSEBtHBHpRDU18P3vmzNvtm5V\nofcnbSr2J0+exOl0Mn78eAAOHjxIYmIiMTExjBkzhsbGRu+x2dnZREdHExsbS3FxsW9Si4jf2bnT\nPBE7cSL86U/Qu7fVieR0bSr2CxcuxOFwYLPZAMjJySExMZGysjJGjx5NTk4OAKWlpaxYsYLS0lLW\nr1/PfffdR3Nzs+/Si4jlDANycyEpCZ5/3jwZ+1WpED/SarGvrq5m7dq1zJgxw7ureVFREenp6QCk\np6ezevVqAAoLC0lNTSU4OJiIiAiioqIoKSnxYXwRsdLRozB1qrnkwfbt8NU//sUPtXqC9qGHHmL+\n/PkcPnzY+1hDQwOhX20fExoaSkNDAwC1tbWMGjXKe5zdbqempuaM75uRkeG973K5cLlc7ckvIhYp\nL4fkZBg+3Nx0RG2bzud2u3G73Z3yXucs9mvWrCEkJASn03nWD7TZbN72ztmeP5PTi72IdC+FhfCj\nH0FmJvzkJ2rb+MrXB8KZmZntfq9zFvvt27dTVFTE2rVrOX78OIcPH2bKlCmEhoZSX19PWFgYdXV1\nhISEABAeHk5VVZX39dXV1YSHh7c7nIj4lxMn4Kmn4JVX4PXXzROy0j2cs2eflZVFVVUVFRUVFBQU\ncMstt/Dyyy+TlJREfn4+APn5+UyYMAGApKQkCgoK8Hg8VFRUUF5eTnx8vO+/ChHxuY8+ApcL9u6F\nt99Woe9uzmue/amWzOOPP86GDRuIiYlh8+bNPP744wA4HA5SUlJwOBz88Ic/JDc395wtHhHpHlav\nhvh4uP12cwmEyy+3OpGcL5txaopNV36ozYYFHysi5+n4cXNZ4jVrYPlyOG3+hVigI7VTyyWIyBmV\nlcFdd8F3vgN79mi1yu5OyyWISAuGAS+8AN/9rjnj5tVXVeh7Ao3sRcSroQFmzIDqanC7IS7O6kTS\nWTSyFxEAVq2CYcNg6FD4299U6HsajexFAtzhw/DTn5qrVP7lL1p7vqfSyF4kgG3ZYo7kL7gA9u1T\noe/JNLIXCUDHj8PcueaVsL//Pdx2m9WJxNdU7EUCzDvvwD33QHS0eV8XSAUGtXFEAsTJkzBvHiQk\nwMMPw2uvqdAHEo3sRQLAv/4F6ekQFAS7d8NVV1mdSLqaRvYiPZhhQF7ev7cL3LRJhT5QaWQv0kM1\nNJhXwH78MbzxBgwebHUisZJG9iI9UGGhuYPU4MFQUqJCLxrZi/QoR47ArFnmSP7VV831bURAI3uR\nHmPHDnM0bxjmBVIq9HI6jexFurkTJ+Dpp82Lo55/3jwRK/J1KvYi3VhZmXmB1KWXmtsFXnGF1YnE\nX6mNI9INGQb87ndmq+a//svc
}
],
"prompt_number": 99
},
{
"cell_type": "code",
"collapsed": false,
"input": [
"# local maxima and minima\n",
"hi = filters.gaussian_filter(filters.maximum_filter(passengers,12),12)\n",
"lo = filters.gaussian_filter(filters.minimum_filter(passengers,12),12)\n",
"plot(lo)\n",
"plot(hi)"
],
"language": "python",
"metadata": {},
"outputs": [
{
"output_type": "pyout",
"prompt_number": 100,
"text": [
"[<matplotlib.lines.Line2D at 0xc625650>]"
]
},
{
"output_type": "display_data",
"png": "iVBORw0KGgoAAAANSUhEUgAAAXsAAAD9CAYAAABdoNd6AAAABHNCSVQICAgIfAhkiAAAAAlwSFlz\nAAALEgAACxIB0t1+/AAAIABJREFUeJzt3Xlc1VX+x/EXCFamqaUsgYWJiBdXVFyqEUU0LQlDKXBB\nJC2dFtubfjOFLYrZNNniTCUqaYmYjuCGuIGae2BWWKCBIgIqiAuoLPf8/viOlKaCCHzv8nk+Hveh\nXu7yBuXN8XzP93xtlFIKIYQQFs1W7wBCCCHqn5S9EEJYASl7IYSwAlL2QghhBaTshRDCCkjZCyGE\nFahR2RcXFzNy5Eg6duyIwWBg165dFBUV4e/vj4eHB4MHD6a4uLjq8TNmzKB9+/Z4enqSlJRUb+GF\nEELUTI3K/vnnn2fYsGEcOHCA/fv34+npSVRUFP7+/mRkZODn50dUVBQA6enpLFmyhPT0dBITE5ky\nZQpGo7FePwkhhBDXV23Znz59mq1btzJhwgQA7OzsaN68OQkJCYSFhQEQFhbGihUrAIiPjyckJAR7\ne3vc3Nxwd3dn9+7d9fgpCCGEqE61ZZ+VlUXr1q0JDw/H29ubiRMnUlJSQkFBAY6OjgA4OjpSUFAA\nwLFjx3B1da16vqurK7m5ufUUXwghRE3YVfeAiooKUlNT+fTTT+nVqxdTp06tmrK5xMbGBhsbm2u+\nxpUfu95jhRBCXFttd7ipdmTv6uqKq6srvXr1AmDkyJGkpqbi5OREfn4+AHl5eTg4OADg4uJCTk5O\n1fOPHj2Ki4vLVQOb+u2tt97SPYPklJzmmtEcclYaK9mXt49hE4cxIX4Cvb/sTev3W2P/tj0OsxwY\nGDNQ94x/vN2Makf2Tk5OtGnThoyMDDw8PNiwYQNeXl54eXkRExPDa6+9RkxMDIGBgQAEBAQQGhrK\niy++SG5uLpmZmfj4+NxUSCGEqCvny8+z7tA6lh9YTuLBRJrf2pxmZ5vxsPPDhHcLx/1Odxxvd7S4\nGYhqyx7gk08+YfTo0ZSVldGuXTvmz59PZWUlwcHBREdH4+bmRlxcHAAGg4Hg4GAMBgN2dnbMmTPH\n4r5oQgjzk5aXxhepXxD3cxxdHLsQ1DGIdwa8w70t7iWyMJIpvaboHbFe1ajsu3btyp49e/50/4YN\nG676+DfeeIM33njj5pKZAF9fX70j1IjkrFvmkNMcMoL+OZVSbM7eTNS2KA6cPMBTPZ5i31P7aNO8\nzWWP0ztnQ7BRNzsRVJs3tbG56fknIYS4FqUU8b/GM33rdM6WneW1+18jtHMojRs11jvaTbmZ7qzR\nyF4IIczF3mN7eWHdC5y9eJZI30gCOgRgayM7w0jZCyEsQu6ZXN7Y9AbrD63nnQHvML7beBrZNtI7\nlsmQH3dCCLNmVEY+3vUxXf/TFZdmLvz6zK9EeEdI0V9BRvZCCLP168lfiUiIwNbGlu0R2/G4y0Pv\nSCZLRvZCCLNTYaxg5raZ3D/vfkI6hZA8PlmKvhoyshdCmJXDxYcZvXw0t9jdwt5Je3Fr4aZ3JLMg\nI3shhNn4Nv1ben3Zi0c7PMr6seul6G+AjOyFECavtLyUF9a9wMbfNrI6dDW9XHrpHcnsyMheCGHS\nDhUdom90X0rKSkh7Kk2Kvpak7IUQJmtt5lr6zevHJO9JLByxkGa3NNM7ktmSaRwhhMkxKiPTt07n\n33v/zbLgZTxwzwN6RzJ7UvZCCJNy+sJpwlaEcaL0BHsm7uHuZnfrHckiyDSOEMJkHDhxAJ+5Ptzd\n7G42h22Woq9DMrIXQpiE9YfWM3r5aGYOmkl493C941gcKXshhO6+/P5L/rH5HywLXsaD9z6odxyL\nJGUvhNCNURl5Y+MbLDuwjC3hW2TLg3okZS+E0MX58vOMWzGO/HP57IjYQasmrfSOZNHkAK0QosEd\nLznOgJgBNG7UmA1jN0jRNwApeyFEgzpw4gB95vZhiPsQFo1YxC12t+gdySrINI4QosFsytpEyLIQ\nZvnPYlzXcXrHsSpS9kKIBjE/bT6vb3ydJSOX4Ovmq3ccqyNlL4SoV0Zl5M3Nb7L4p8WkjE/Bs5Wn\n3pGskpS9EKLeXKi4QHh8OIeLD7MzYietb2+tdySrJQdohRD14kTJCfy+8sOojGwct1GKXmdS9kKI\nOvfryV/pG92X/vf2Z3HQYm6zv03vSFZPpnGEEHUqJTuF4G+DmeE3gwndJ+gdR/yPlL0Qos4s/GEh\nLyW9xOKgxfjd56d3HPEHUvZCiJumlGJayjRifogheXwyhtYGvSOJK0jZCyFuysWKizy58kkyCjPY\nGbETx6aOekcSVyEHaIUQtVZYWoj/Qn/Ol59nc9hmKXoTJmUvhKiVg0UH6Rvdlz6ufYgbFUcT+yZ6\nRxLXIWUvhLhh245s44F5D/BS35d43/99bG2kSkydzNkLIW7I4h8X83zi8ywcsZAh7kP0jiNqSMpe\nCFEjSineTnmb+fvms3HcRjo7dtY7krgBUvZCiGpdqLhAREIEB4sOsvPJnTg1ddI7krhBNZpoc3Nz\no0uXLnTv3h0fHx8AioqK8Pf3x8PDg8GDB1NcXFz1+BkzZtC+fXs8PT1JSkqqn+RCiAZxvOQ4fl/5\nUWGsIDksWYreTNWo7G1sbEhOTiYtLY3du3cDEBUVhb+/PxkZGfj5+REVFQVAeno6S5YsIT09ncTE\nRKZMmYLRaKy/z0AIUW/ST6TTZ24fBrYdKHvcmLkaH0JXSl3254SEBMLCwgAICwtjxYoVAMTHxxMS\nEoK9vT1ubm64u7tX/YAQQpiP9YfW47vAl0jfSN4Z8I6suDFzNZqzt7GxYdCgQTRq1IinnnqKiRMn\nUlBQgKOjdgKFo6MjBQUFABw7dow+ffpUPdfV1ZXc3Nw/vWZkZGTV7319ffH19b2JT0MIUZf+s/c/\nRCZHsix4GQ/e+6DecaxWcnIyycnJdfJaNSr77777DmdnZ06cOIG/vz+enpdfacbGxgYbG5trPv9q\nH/tj2QshTEOlsZKX17/M2sy1bJuwDfc73fWOZNWuHAhPmzat1q9Vo7J3dnYGoHXr1owYMYLdu3fj\n6OhIfn4+Tk5O5OXl4eDgAICLiws5OTlVzz169CguLi61DiiEaBhnL54ldHkopeWl7IjYQcvbWuod\nSdShaifhSktLOXv2LAAlJSUkJSXRuXNnAgICiImJASAmJobAwEAAAgICiI2NpaysjKysLDIzM6tW\n8AghTFPO6RwemP8Azk2dSRydKEVvgaod2RcUFDBixAgAKioqGD16NIMHD6Znz54EBwcTHR2Nm5sb\ncXFxABgMBoKDgzEYDNjZ2TFnzpzrTvEIIfS1J3cPgUsCebHPi7zY90X5frVQNurKZTYN8aY2Nn9a\n3SOEaHjfpn/L5NWT+XL4lwR6BuodR1TjZrpTzqAVwgoppXhv63t88f0XrBuzDm9nb70jiXomZS+E\nlbm09UFGYQY7n9zJ3c3u1juSaAByloQQVqTgXAEDYgZQYawgZXyKFL0VkbIXwkrsL9hP77m9GdJu\nCLFBsXKxESsj0zhCWIGVv65kQsIEPhn6CU90ekLvOEIHUvZCWDClFP/c8U/+tfNfrApZRW/X3npH\nEjqRshfCQpVVljF59WS+P/Y9OyJ2cE/ze/SOJHQkZS+EBTpZepKguCBa3tqSbRO20bRxU70jCZ3J\nAVohLMyBEwfoM7cPfVz7sPzx5VL0ApCRvRAWJelQEmOWj+F9//cZ32283nGECZGyF8JCfLr7U97d\n8q7sQS+uSspeCDNXYazg+cTnSc5OZnvEdu5reZ/ekYQJkrIXwowVXyhm1NJR2NnasX3Cdprf2lzv\nSMJEyQFaIczUwaKD9JnbB6/WXqwMWSlFL65Lyl4IM5ScncwD8x7ghT4v8NFDH2FnK/9JF9cn/0KE\nMDPz0+bz+sbX+eaxb/C7z0/vOMJMSNkLYSaMysibm99k8U+LSRmfgmcrT70jCTMiZS+EGbhQcYHw\n+HAOFx9mZ8ROWt/eWu9IwszI
}
],
"prompt_number": 100
},
{
"cell_type": "code",
"collapsed": false,
"input": [
"# detrending\n",
"variation = (passengers-lo)/(hi-lo)\n",
"plot(variation)"
],
"language": "python",
"metadata": {},
"outputs": [
{
"output_type": "pyout",
"prompt_number": 101,
"text": [
"[<matplotlib.lines.Line2D at 0xca36750>]"
]
},
{
"output_type": "display_data",
"png": "iVBORw0KGgoAAAANSUhEUgAAAYAAAAD9CAYAAAC1DKAUAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nZvczD2gVgLSA+EMqAAOoqxCK
}
],
"prompt_number": 101
},
{
"cell_type": "code",
"collapsed": false,
"input": [
"# discovering periodicity by autocorrelation\n",
"ac = correlate(variation,variation,mode='full')\n",
"ac = ac[len(ac)//2:]\n",
"plot(ac)"
],
"language": "python",
"metadata": {},
"outputs": [
{
"output_type": "pyout",
"prompt_number": 102,
"text": [
"[<matplotlib.lines.Line2D at 0xca657d0>]"
]
},
{
"output_type": "display_data",
"png": "iVBORw0KGgoAAAANSUhEUgAAAXUAAAD9CAYAAABDaefJAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz\nAAALEgAACxIB0t1+/AAAIABJREFUeJzt3Xt4VNXVP/DvBAKIyFWSAAEJSCAhIQyg+Cr+DKQBkYsI\nFBHByMUL1laNrwXa1xr6KkR5KUW0LbZYg7SirUVAMYJCRFFACEHuQZxoICFcQgJBMCTZvz+Wk0yS\nuZxz5sxkJnw/z8NjMnNmzy59WLNnnbXXtiilFIiIqFEIaegJEBGReRjUiYgaEQZ1IqJGhEGdiKgR\nYVAnImpEGNSJiBoRTUG9srISVqsVY8aMAQCkpaUhMjISVqsVVqsVmZmZPp0kERFp01TLRUuXLkVs\nbCwuXLgAALBYLEhNTUVqaqpPJ0dERPp4XKkfP34cGzZswKxZs2Dfp6SUAvcsEREFHo9B/amnnsKi\nRYsQElJzqcViwbJly5CQkICZM2eipKTEp5MkIiKNlBvr169Xjz32mFJKqS1btqjRo0crpZQqKipS\nVVVVqqqqSv32t79VM2bMcPp6APzDP/zDP/xj4I9Rbl85b948FRkZqbp3764iIiJUy5Yt1bRp02pd\nY7PZVFxcnMugHgyee+65hp6CR8EwR6U4T7NxnuYKlnl6Ezvdpl8WLFiA/Px82Gw2rF69GsOGDcPK\nlStRWFhYfc2aNWsQHx/vbhgiIvITTdUvAKCUgsViAQD8+te/xt69e2GxWBAVFYXly5f7bIJERKSd\n5qCemJiIxMREAMCbb77pq/k0CPv/rkAWDHMEOE+zcZ7mCpZ5esPyU/7GN4NbLCx9JCLSyZvYyTYB\nRESNCIM6EVEjwqBORNSIMKgTETUiDOpERI0IgzoRUSPCoE5E1IgwqBMRNSIM6kREjQiDOhFRI8Kg\nTkTUiDCoExE1IgzqRESNCIM6EVEjwqBORNSIaArqlZWVsFqtGDNmDACguLgYycnJiI6OxvDhw1FS\nUuLTSRIRkTaagvrSpUsRGxtbfZxdeno6kpOTkZubi6SkJKSnp/t0kkREpI3HoH78+HFs2LABs2bN\nqj6JY926dUhJSQEApKSk4L333nP5+itXTJopERF55PGM0qeeegqLFi3C+fPnqx8rKipCeHg4ACA8\nPBxFRUUuXz9nThpat5afHc85JSIikZWVhaysLFPGchvU33//fYSFhcFqtbp8Q4vFUp2Wcebee9Mw\neLBXcyQiatTqLnjnz59veCy3Qf2LL77AunXrsGHDBly+fBnnz5/HtGnTEB4ejpMnTyIiIgKFhYUI\nCwtzOcaJE4bnRkREOrnNqS9YsAD5+fmw2WxYvXo1hg0bhjfffBNjx45FRkYGACAjIwPjxo1zOQaD\nOhGR/+iqU7enWebOnYtNmzYhOjoamzdvxty5c12+hkGdiMh/LMpe0uKLwS0WTJumsHKlr96BiKjx\nsVgsMBqafb6jlCt1IiL/YVAnImpEGNSJiBoRnwd1pQCHfUtERORDPg/qnTtztU5E5C8+D+pdugAF\nBb5+FyIiAvy0UmdQJyLyD58H9Q4dgOJi34y9dCnw2We+GZuIKBj5PKi3b++7oL52LYM6EZGjoF6p\n22xAXp5vxiYiCkZ+WamfPWv+uBUVQH6+BHYiIhJBm37Jz5f/cqVORFQjaIO6zQYMHCjBvarK/PGJ\niIJR0ObUbTYgJkY+NFgySUQkgjanbrMBUVFA9+5MwRAR2fk8qLdpA1y4AFRWmjuuPahHRTGoExHZ\neQzqly9fxuDBg9G/f3/ExsZi3rx5AIC0tDRERkbCarXCarUiMzPT6eubNJHAXlJi7sQdV+qsgCEi\nEm4PngaAFi1aYMuWLWjZsiUqKiowZMgQfP7557BYLEhNTUVqaqrHN7GnYDp0MGXOAGoH9e3bzRuX\niCiYaUq/tGzZEgBQXl6OyspKtGvXDgA0H7dkdgXMpUvAuXPSV4bpFyKiGh5X6gBQVVWFAQMG4Nix\nY5g9ezb69u2Lf//731i2bBlWrlyJQYMGYfHixWjbtm2916alpaG4GFi2DLh8ORGJiYleTzovD+jW\nDQgJYfqFiIJfVlYWsrKyTBlL18HTpaWlGDFiBNLT0xEbG4uOHTsCAJ599lkUFhZixYoVtQf/6fDU\nqVOBO+8Epk41Zc7YsEGaeX30EfDjj0Dr1sDFi0BTTR9RRESBzW8HT7dp0wajRo3Crl27EBYWBovF\nAovFglmzZmHnzp0uX2d2WaM9nw4AzZsDYWE8iIOICNAQ1M+cOYOSn0pXLl26hE2bNsFqteLkyZPV\n16xZswbx8fEuxzA7p+4Y1AHWqhMR2XlMWBQWFiIlJQVVVVWoqqrCtGnTkJSUhAceeAA5OTmwWCyI\niorC8uXLXY7Rvj1w9Kh5k7bZgJtvrvndfrP0jjvMew+7jz4CmjUDhg41f2wiIrN5DOrx8fHIzs6u\n9/jKlSs1v0mHDuaWHRYWSuWLXdeuNQ2+zPbWW0DLlgzqRBQc/HJr0ez0y5kzwE/3aAHIz75Kv+Tm\nAtde65uxiYjM5vM2AYBvgvr119f8fv31wOnT5o3vKDfX3NQREZEv+SWom9mpsaICOH8ecCyJv/56\nCfRmKy4GysuBkyeBy5fNH5+IyGx+W6mbVdJYXAy0ayc9Zew6dvRNUD96FOjdG7jhBuDbb80fn4jI\nbH4J6mZ2ajx9unY+HfBd+iU3F4iOBnr1kp+JiAKdX26UNmkiuz5LSrxv6lU3nw74Lv2SmysBvbSU\neXUiCg5+WakD5uXVnQV1e3XKDz94P74jx5U6gzoRBQO/BXWz8urOgjrgmxTM0aMS1KOjGdSJKDj4\nNaibsVJ3llMHzE/BKFWTfuFKnYiCRaNIvwDmV8AUFkpap00b2bF69qz56R0iIrMx/eKCPZ8OSN/2\nHj2Ab74xb3wiIl/wW1Dv2NGcoFu3RYCd2ekXez7djmWNRBQM/BbUw8OBoiLvx/FX+sVxpQ4wr05E\nwcFvQT0sDDh1yvtxTp/2T/olP1+OzLPr0YM924ko8AVdUHeXUzdzpX7qlHy7sIuIMOebBhGRLwVV\n+uWHH4CqKuetcM0O6kVF8kFkZ1b6yJmMDCA11TdjE9HVxW1Qv3z5MgYPHoz+/fsjNjYW8+bNAwAU\nFxcjOTkZ0dHRGD58ePVxd+6YsVI/e1aCt8VS/zmzbsTanTrlv6D+1VfmHiJCRFcvt0G9RYsW2LJl\nC3JycvD1119jy5Yt+Pzzz5Geno7k5GTk5uYiKSkJ6enpHt+oVStp6HXxovHJusqnA+au1Csq6vep\nCQvzXVA/fBg4cEA2PBERecNj+qVly5YAgPLyclRWVqJdu3ZYt24dUlJSAAApKSl47733PL6RxSKr\nXW9W667y6YDUwZ87J+kZb509W7+9b6tWEnTLyrwfv65Dh6Rf+/Hj5o9NRFcXj10aq6qqMGDAABw7\ndgyzZ89G3759UVRUhPCf7iKGh4ejyM0SNi0trfrnFi0ScepUIqKijE3WVY06AISGAtddJ4Hd206Q\ndVMvQM2HUlGRBHiznD8v3wqGDAH275fdq0R0dcnKykJWVpYpY3kM6iEhIcjJyUFpaSlGjBiBLVu2\n1HreYrHA4izJ/RPHoL5rl+9W6kBNCsbboF5UVLvyxc4e1Hv29G58R0eOSD18v36Sghk50ryxiSg4\nJCYmIjExsfr3+fPnGx5Lc/VLmzZtMGrUKOzevRvh4eE4efIkAKCwsBBhdZe1Lnh7s9FdTh0wL6/u\nbKUO+OZm6aFDQEwM0LevrNSJiLzhNqifOXOmurLl0qVL2LRpE6xWK8aOHYuMjAwAQEZGBsaNG6fp\nzbytgHGXfgHMq4DxZ1A/fBjo0weIi2NQJyLvuU2/FBYWIiUlBVVVVaiqqsK0adOQlJQEq9WKSZMm\nYcWKFejevTveeecdTW8WFgZ8953xyWpNv3jL30H9vvuA2FhZtVdVSQMxIiIj3Ab1+Ph4ZGdn13u8\nffv2+Pjjj3W/WXg4sHOn7pdV
}
],
"prompt_number": 102
},
{
"cell_type": "code",
"collapsed": false,
"input": [
"# discovering periodicity by Fourier transform\n",
"plot(abs(fft.fft(variation-mean(variation)))[:40])"
],
"language": "python",
"metadata": {},
"outputs": [
{
"output_type": "pyout",
"prompt_number": 103,
"text": [
"[<matplotlib.lines.Line2D at 0xce13f90>]"
]
},
{
"output_type": "display_data",
"png": "iVBORw0KGgoAAAANSUhEUgAAAXIAAAD9CAYAAAChtfywAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nAE6nE2VlZaYWRcYxyO2tfUfu
}
],
"prompt_number": 103
},
{
"cell_type": "heading",
"level": 1,
"metadata": {},
"source": [
"Discrete Processes"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"- exogenous input\n",
"- dependence on past outputs\n",
"- noise"
]
},
{
"cell_type": "code",
"collapsed": false,
"input": [
"# random walk\n",
"ys = []\n",
"y = 0.0\n",
"for i in range(1000):\n",
" ys.append(y)\n",
" y = y+randn()\n",
"ys = array(ys)\n",
"plot(ys)"
],
"language": "python",
"metadata": {},
"outputs": [
{
"output_type": "pyout",
"prompt_number": 108,
"text": [
"[<matplotlib.lines.Line2D at 0xd7a3c10>]"
]
},
{
"output_type": "display_data",
"png": "iVBORw0KGgoAAAANSUhEUgAAAYAAAAD9CAYAAAC1DKAUAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nbbWks5QUOtbUqL/XYCAroK6O
}
],
"prompt_number": 108
},
{
"cell_type": "code",
"collapsed": false,
"input": [
"# discovering periodicity by autocorrelation\n",
"ac = correlate(ys,ys,mode='full')\n",
"ac = ac[len(ac)//2:]\n",
"plot(ac)"
],
"language": "python",
"metadata": {},
"outputs": [
{
"output_type": "pyout",
"prompt_number": 109,
"text": [
"[<matplotlib.lines.Line2D at 0xd7d1550>]"
]
},
{
"output_type": "display_data",
"png": "iVBORw0KGgoAAAANSUhEUgAAAZMAAAD9CAYAAAB5lZr/AAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\niIhceH7tGjxgwAAeeughrrzy
}
],
"prompt_number": 109
},
{
"cell_type": "code",
"collapsed": false,
"input": [
"# random walk with uniform increments\n",
"ys = []\n",
"y = 0.0\n",
"for i in range(1000):\n",
" ys.append(y)\n",
" y = y+rand()-0.5\n",
"plot(ys)"
],
"language": "python",
"metadata": {},
"outputs": [
{
"output_type": "pyout",
"prompt_number": 113,
"text": [
"[<matplotlib.lines.Line2D at 0xe4d5410>]"
]
},
{
"output_type": "display_data",
"png": "iVBORw0KGgoAAAANSUhEUgAAAYAAAAD9CAYAAAC1DKAUAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nVWoIlZUyOcvf9NwBPfSzoSIj
}
],
"prompt_number": 113
},
{
"cell_type": "code",
"collapsed": false,
"input": [
"# random walk with long-tailed increments\n",
"from scipy import stats\n",
"ys = []\n",
"y = 0.0\n",
"for i in range(1000):\n",
" ys.append(y)\n",
" y = y+stats.cauchy.rvs()\n",
"plot(ys)"
],
"language": "python",
"metadata": {},
"outputs": [
{
"output_type": "pyout",
"prompt_number": 153,
"text": [
"[<matplotlib.lines.Line2D at 0x13b63890>]"
]
},
{
"output_type": "display_data",
"png": "iVBORw0KGgoAAAANSUhEUgAAAYYAAAD9CAYAAAC4EtBTAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\njRthwwaz0F90tFnP6ZFHmq7M
}
],
"prompt_number": 153
},
{
"cell_type": "code",
"collapsed": false,
"input": [
"# autoregressive process (no noise)\n",
"ys = [0.1,0.5,1.0]\n",
"for i in range(1000):\n",
" y = ys[-1]+0.01*ys[-2]-0.02*ys[-3]\n",
" ys.append(y)\n",
"plot(ys)"
],
"language": "python",
"metadata": {},
"outputs": [
{
"output_type": "pyout",
"prompt_number": 120,
"text": [
"[<matplotlib.lines.Line2D at 0xf250790>]"
]
},
{
"output_type": "display_data",
"png": "iVBORw0KGgoAAAANSUhEUgAAAXsAAAD9CAYAAABdoNd6AAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nVNiDxu1FRK5ExEy9vKihAVJS
}
],
"prompt_number": 120
},
{
"cell_type": "code",
"collapsed": false,
"input": [
"# autoregressive process with noise\n",
"ys = zeros(1000)\n",
"params = array([0.2,0.3,0.35])\n",
"for i in range(4,1000):\n",
" y = dot(params,ys[i-4:i-1])+0.01*randn()\n",
" ys[i] = y\n",
"plot(ys[:100])"
],
"language": "python",
"metadata": {},
"outputs": [
{
"output_type": "pyout",
"prompt_number": 155,
"text": [
"[<matplotlib.lines.Line2D at 0x13d859d0>]"
]
},
{
"output_type": "display_data",
"png": "iVBORw0KGgoAAAANSUhEUgAAAYYAAAD9CAYAAAC4EtBTAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nf8azQD7ITW5mmxGGqipg5Uqa
}
],
"prompt_number": 155
},
{
"cell_type": "code",
"collapsed": false,
"input": [
"# moving average process with noise\n",
"ys = zeros(1000)\n",
"ns = zeros(1000)\n",
"params = array([0.1,0.1,0.1])\n",
"for i in range(4,1000):\n",
" y = dot(params,ns[i-4:i-1])+randn()\n",
" ys[i] = y\n",
"plot(ys[:100])"
],
"language": "python",
"metadata": {},
"outputs": [
{
"output_type": "pyout",
"prompt_number": 149,
"text": [
"[<matplotlib.lines.Line2D at 0x12be3f90>]"
]
},
{
"output_type": "display_data",
"png": "iVBORw0KGgoAAAANSUhEUgAAAXYAAAD9CAYAAACoXlzKAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nc3bxRCfEVVFCT42y6e2NbhSj
}
],
"prompt_number": 149
},
{
"cell_type": "heading",
"level": 1,
"metadata": {},
"source": [
"Parameter Estimation"
]
},
{
"cell_type": "code",
"collapsed": false,
"input": [
"# autoregressive process with noise\n",
"ys = zeros(10000)\n",
"params = array([0.2,0.3,0.35])\n",
"for i in range(4,10000):\n",
" y = dot(params,ys[i-4:i-1])+0.01*randn()\n",
" ys[i] = y\n",
"plot(ys[:100])"
],
"language": "python",
"metadata": {},
"outputs": [
{
"output_type": "pyout",
"prompt_number": 234,
"text": [
"[<matplotlib.lines.Line2D at 0x1b4e07d0>]"
]
},
{
"output_type": "display_data",
"png": "iVBORw0KGgoAAAANSUhEUgAAAU8AAAE1CAYAAABndGJVAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nJLkrep58kv3LBRfQdp6A8aCR
}
],
"prompt_number": 234
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"We now attempt to predict the value at time $t$ as a linear combination of values at times $t-1$ through $t-4$. First, let us generate a training set.\n"
]
},
{
"cell_type": "code",
"collapsed": false,
"input": [
"data = array([ys[i-4:i-1] for i in range(4,10000)])\n",
"pred = array([ys[i] for i in range(4,10000)])"
],
"language": "python",
"metadata": {},
"outputs": [],
"prompt_number": 235
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"Now let us perform linear fitting.\n",
"\n"
]
},
{
"cell_type": "code",
"collapsed": false,
"input": [
"from numpy import linalg\n",
"x,residuals,rank,svs = linalg.lstsq(data,pred)\n",
"x"
],
"language": "python",
"metadata": {},
"outputs": [
{
"output_type": "pyout",
"prompt_number": 236,
"text": [
"array([ 0.22427482, 0.29612102, 0.33520334])"
]
}
],
"prompt_number": 236
},
{
"cell_type": "code",
"collapsed": false,
"input": [
"roots(x)"
],
"language": "python",
"metadata": {},
"outputs": [
{
"output_type": "pyout",
"prompt_number": 237,
"text": [
"array([-0.66017447+1.02897009j, -0.66017447-1.02897009j])"
]
}
],
"prompt_number": 237
}
],
"metadata": {}
}
]
}