2models_kancolle.ipynb

{
 "metadata": {
  "name": "",
  "signature": "sha256:6262d7b83e1145a1522ba20a5d73a4c166deffd1622415d84c13c8fc7677417f"
 },
 "nbformat": 3,
 "nbformat_minor": 0,
 "worksheets": [
  {
   "cells": [
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "import numpy as np\n",
      "import pandas as pd\n",
      "import matplotlib.pylab as plt\n",
      "%matplotlib inline\n",
      "from datetime import datetime\n",
      "from datetime import timedelta\n",
      "\n",
      "import pickle as pkl"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [],
     "prompt_number": 34
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "import NiconicoSnapshotAPIWrapper as nico\n",
      "api = nico.NiconicoSnapshotAPIWrapper('NiconicoSnapshotAPIWrapper')"
     ],
     "language": "python",
     "metadata": {},
     "outputs": []
    },
    {
     "cell_type": "heading",
     "level": 2,
     "metadata": {},
     "source": [
      "\u30c7\u30fc\u30bf\u306e\u53d6\u5f97"
     ]
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "et=datetime(2015,4,1)\n",
      "N=201\n",
      "daterange=[ ( (et + timedelta(days=t)).strftime(\"%Y-%m-%d %H:%M:%S\"),(et + timedelta(days=t+1)).strftime(\"%Y-%m-%d %H:%M:%S\")) for t in xrange(N)]\n",
      "startdate=zip(*daterange)[0]\n",
      "print daterange[-1]"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": [
        "('2015-10-18 00:00:00', '2015-10-19 00:00:00')\n"
       ]
      }
     ],
     "prompt_number": 73
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "num=[api.query(u\"(\u8266\u3053\u308c or \u8266\u968a\u3053\u308c\u304f\u3057\u3087\u3093) \", retry = 3, size = 0, filters=[api.makeFilterRange('start_time', d[0],d[1])]).total for d in daterange]"
     ],
     "language": "python",
     "metadata": {},
     "outputs": []
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "import pystan\n",
      "pystan.__version__"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "metadata": {},
       "output_type": "pyout",
       "prompt_number": 6,
       "text": [
        "'2.8.0.0'"
       ]
      }
     ],
     "prompt_number": 6
    },
    {
     "cell_type": "heading",
     "level": 2,
     "metadata": {},
     "source": [
      "Poisson model"
     ]
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "model0=\"\"\"\n",
      "data{\n",
      "  int<lower=1> N;\n",
      "  int<lower=1> N_next;\n",
      "  int<lower=0,upper=1000> X[N];\n",
      "}\n",
      "\n",
      "parameters {\n",
      "  real trend[N];\n",
      "  real mm[N];\n",
      "  real<lower=0,upper=20> s_trend;\n",
      "  real<lower=0,upper=20> s_mm;\n",
      "  real<lower=0,upper=20> s_g;\n",
      "}\n",
      "model{\n",
      "  real m[N];\n",
      "  \n",
      " for(i in 3:N)\n",
      "   trend[i]~normal(2*trend[i-1]-trend[i-2],s_trend);\n",
      "\n",
      "  for(i in 7:N)\n",
      "    mm[i]~normal(-mm[i-1]-mm[i-2]-mm[i-3]-mm[i-4]-mm[i-5]-mm[i-6],s_mm);\n",
      "\n",
      "  for(i in 1:N)\n",
      "        m[i]<-trend[i]+mm[i];\n",
      "\n",
      " for(i in 1:N)\n",
      "      X[i]~poisson_log(m[i]);\n",
      "}\n",
      "generated quantities {\n",
      "   real mm_all[N+N_next];\n",
      "   real trend_all[N+N_next];\n",
      "   real m_next[N_next];\n",
      "   real y_next[N_next];\n",
      "\n",
      "   for (t in 1:N){\n",
      "      mm_all[t] <- mm[t];\n",
      "      trend_all[t] <- trend[t];\n",
      "   }\n",
      "   for (t in (N+1):(N+N_next)){\n",
      "      mm_all[t] <- normal_rng(mm_all[t-1], s_mm);\n",
      "      trend_all[t] <- normal_rng(-trend_all[t-1]-trend_all[t-2]-trend_all[t-3], s_trend);\n",
      "   }\n",
      "\n",
      "   for (t in 1:N_next){\n",
      "      m_next[t]<-trend[t]+mm[t];\n",
      "      y_next[t] <- poisson_log_rng(m_next[t]);\n",
      "   }\n",
      "}\n",
      "\n",
      "\"\"\""
     ],
     "language": "python",
     "metadata": {},
     "outputs": [],
     "prompt_number": 15
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "with open(\"logpoissonmodel.stan\",\"w\") as fp:\n",
      "    fp.write(model0)"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [],
     "prompt_number": 30
    },
    {
     "cell_type": "heading",
     "level": 2,
     "metadata": {},
     "source": [
      "\u6b63\u898f\u5206\u5e03\u30e2\u30c7\u30eb"
     ]
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "model2=\"\"\"\n",
      "data{\n",
      "  int<lower=1> N;\n",
      "  int<lower=1> N_next;\n",
      "  real <lower=0,upper=1000> X[N];\n",
      "}\n",
      "\n",
      "parameters {\n",
      "  real trend[N];\n",
      "  real mm[N];\n",
      "  real<lower=0,upper=20> s_trend;\n",
      "  real<lower=0,upper=20> s_mm;\n",
      "  real<lower=0,upper=20> s_g;\n",
      "}\n",
      "model{\n",
      "  real m[N];\n",
      "  \n",
      " for(i in 3:N)\n",
      "   trend[i]~normal(2*trend[i-1]-trend[i-2],s_trend);\n",
      "\n",
      "  for(i in 7:N)\n",
      "    mm[i]~normal(-mm[i-1]-mm[i-2]-mm[i-3]-mm[i-4]-mm[i-5]-mm[i-6],s_mm);\n",
      "\n",
      "  for(i in 1:N)\n",
      "        m[i]<-trend[i]+mm[i];\n",
      "\n",
      " for(i in 1:N)\n",
      "      X[i]~normal(m[i],s_g);\n",
      "}\n",
      "\n",
      "generated quantities {\n",
      "   real mm_all[N+N_next];\n",
      "   real trend_all[N+N_next];\n",
      "   real m_next[N_next];\n",
      "   real y_next[N_next];\n",
      "\n",
      "   for (t in 1:N){\n",
      "      mm_all[t] <- mm[t];\n",
      "      trend_all[t] <- trend[t];\n",
      "   }\n",
      "   for (t in (N+1):(N+N_next)){\n",
      "      mm_all[t] <- normal_rng(mm_all[t-1], s_mm);\n",
      "      trend_all[t] <- normal_rng(-trend_all[t-1]-trend_all[t-2]-trend_all[t-3], s_trend);\n",
      "   }\n",
      "\n",
      "   for (t in 1:N_next){\n",
      "      m_next[t]<-trend[t]+mm[t];\n",
      "      y_next[t] <- normal_rng(m_next[t],s_g);\n",
      "   }\n",
      "}\n",
      "\n",
      "\"\"\""
     ],
     "language": "python",
     "metadata": {},
     "outputs": [],
     "prompt_number": 67
    },
    {
     "cell_type": "heading",
     "level": 2,
     "metadata": {},
     "source": [
      "data"
     ]
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "num=np.loadtxt(\"kancolle_201510.csv\",dtype=np.int)\n",
      "num_d=np.loadtxt(\"kancolle_201510.csv\")\n",
      "N=40\n",
      "ndat=num[40:N+40]\n",
      "ndatdates=idates[40:N+40]\n",
      "dd={\"N\":N,\"N_next\":10,\"X\":ndat}\n",
      "\n",
      "idates=zip(*daterange)[0]\n",
      "#idates=pd.date_range('1/4/2015', periods=N)"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [],
     "prompt_number": 80
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "nc=pd.TimeSeries(data=num,index=idates)\n",
      "nc.plot(figsize=(10,10))"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "metadata": {},
       "output_type": "pyout",
       "prompt_number": 77,
       "text": [
        "<matplotlib.axes.AxesSubplot at 0x116b866d0>"
       ]
      },
      {
       "metadata": {},
       "output_type": "display_data",
       "png": "iVBORw0KGgoAAAANSUhEUgAAAqUAAAJPCAYAAABFB63WAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nCCGEEEIIIYQQQgghhBBCCCGEkDD4fzcJLxfNBBy4AAAAAElFTkSuQmCC\n",
       "text": [
        "<matplotlib.figure.Figure at 0x116b86f10>"
       ]
      }
     ],
     "prompt_number": 77
    },
    {
     "cell_type": "markdown",
     "metadata": {},
     "source": [
      "\u3068\u308a\u3042\u3048\u305a40\u65e5\u5206\u3060\u3051\u53d6\u308a\u51fa\u3059"
     ]
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "pd.TimeSeries(data=ndat,index=ndatdates).plot(figsize=(20,10))"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "metadata": {},
       "output_type": "pyout",
       "prompt_number": 81,
       "text": [
        "<matplotlib.axes.AxesSubplot at 0x116b81590>"
       ]
      },
      {
       "metadata": {},
       "output_type": "display_data",
       "png": "iVBORw0KGgoAAAANSUhEUgAABJ8AAAJPCAYAAADSY3VkAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nrdyldxGXAAAAAElFTkSuQmCC\n",
       "text": [
        "<matplotlib.figure.Figure at 0x1147c8290>"
       ]
      }
     ],
     "prompt_number": 81
    },
    {
     "cell_type": "heading",
     "level": 3,
     "metadata": {},
     "source": [
      "\u6b63\u898f\u5206\u5e03\u30e2\u30c7\u30eb\u306e\u8a08\u7b97"
     ]
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "stanmodel = pystan.StanModel(model_code=model2)\n",
      "fit = stanmodel.sampling(data=dd, iter=10200, warmup=200, thin=10, chains=3, seed=71)\n",
      "print fit\n",
      "res_norm=fit.extract()\n",
      "\n",
      "with open(\"model_normal40\",\"wb\") as fpp:\n",
      "    pkl.dump(res_norm,fpp)"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "output_type": "stream",
       "stream": "stderr",
       "text": [
        "INFO:pystan:COMPILING THE C++ CODE FOR MODEL anon_model_6a587763fbc889ffde98580835aadde7 NOW.\n"
       ]
      },
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": [
        "Inference for Stan model: anon_model_6a587763fbc889ffde98580835aadde7.\n",
        "3 chains, each with iter=10200; warmup=200; thin=10; \n",
        "post-warmup draws per chain=1000, total post-warmup draws=3000.\n",
        "\n",
        "                mean se_mean     sd   2.5%    25%    50%    75%  97.5%  n_eff   Rhat\n",
        "trend[0]       138.2    0.34  10.26 116.49  131.9 138.62 145.23 157.06  900.0    1.0\n",
        "trend[1]      136.93    0.27   8.23 119.72 131.74 137.09 142.55 152.52  931.0    1.0\n",
        "trend[2]      135.32    0.23   7.07  120.7 130.75 135.41 139.99 149.25  923.0    1.0\n",
        "trend[3]      133.44    0.22   6.56 120.22 129.06 133.52  137.5 146.69  871.0    1.0\n",
        "trend[4]      131.39    0.23    6.4 119.18 127.29 131.28 135.38 144.51  766.0    1.0\n",
        "trend[5]      129.35    0.24   6.42 117.08 125.15 129.13 133.43  142.8  706.0    1.0\n",
        "trend[6]      127.06    0.25   6.51 114.97 122.69 126.51 131.06 140.73  677.0    1.0\n",
        "trend[7]      124.19    0.24   6.47 112.27 119.88 123.84 127.95 138.79  703.0    1.0\n",
        "trend[8]      120.54    0.21   6.02  109.2 116.61 120.37 123.93  133.7  811.0    1.0\n",
        "trend[9]      116.62    0.18   5.58 105.99 112.94  116.6 120.18  127.8  917.0    1.0\n",
        "trend[10]     112.65    0.21    5.5 101.63 109.04 112.79 116.37 122.93  685.0    1.0\n",
        "trend[11]     108.92    0.25   5.75  97.03  105.3 109.13 112.66 119.49  541.0   1.01\n",
        "trend[12]     105.43    0.35   6.22  92.34  101.6 105.72 109.49 118.22  315.0   1.01\n",
        "trend[13]     102.51     0.4   6.52  88.61  98.52 102.87 106.76 116.34  271.0   1.01\n",
        "trend[14]     100.31    0.41   6.53  86.42  96.24 100.54 104.54 113.61  252.0   1.01\n",
        "trend[15]      98.81    0.41   6.31  85.76  94.86  98.95 102.83 111.94  239.0   1.01\n",
        "trend[16]      97.78    0.39   6.07  85.57  93.79  97.84  101.5 111.53  240.0   1.01\n",
        "trend[17]      97.03    0.38   6.05  85.63  92.95  97.05 100.62  112.0  256.0   1.01\n",
        "trend[18]      96.35    0.37   6.12  84.94  92.17  96.14  99.91 111.18  276.0   1.01\n",
        "trend[19]      95.39    0.37   6.08   84.1  91.24  95.22  99.11 109.93  275.0   1.01\n",
        "trend[20]      94.33    0.35   5.96  82.99  90.37  94.28  98.01 108.45  284.0   1.01\n",
        "trend[21]       93.4    0.37   5.94  82.02  89.42  93.46   97.1 107.58  255.0   1.01\n",
        "trend[22]      92.79    0.38   5.91  81.36  88.92  92.83   96.5 106.61  246.0   1.01\n",
        "trend[23]      92.39    0.37   5.98  80.73  88.49  92.45  96.09 106.23  267.0   1.01\n",
        "trend[24]      92.06    0.36   5.99  80.29  88.25  92.16  95.75 106.34  278.0   1.01\n",
        "trend[25]      91.78    0.35   5.98  80.01  88.03  91.93   95.5  105.9  292.0   1.01\n",
        "trend[26]      91.83    0.32   5.92  80.04   88.1  91.73  95.55 104.89  333.0   1.01\n",
        "trend[27]      92.21    0.29   5.87  80.34  88.45  92.13  95.95 104.17  410.0   1.01\n",
        "trend[28]      92.83    0.25   5.78  81.32  89.02  92.67  96.65 103.62  540.0    1.0\n",
        "trend[29]      93.55    0.21   5.73  81.88  89.85   93.4  97.38 103.98  752.0    1.0\n",
        "trend[30]      94.47    0.21   5.72  83.65  90.63  94.32  98.24 105.37  767.0    1.0\n",
        "trend[31]      95.53    0.22   5.83  84.57  91.49  95.33  99.48 107.41  712.0    1.0\n",
        "trend[32]      96.52    0.25   6.16  85.25  92.26  96.28 100.18 109.59  588.0    1.0\n",
        "trend[33]      97.06    0.27   6.46  85.95   92.6  96.72 100.76 111.07  552.0   1.01\n",
        "trend[34]      96.95    0.27   6.45  85.73  92.55  96.59  101.0 110.45  562.0   1.01\n",
        "trend[35]      96.28    0.25   6.37  84.59  91.94  95.84 100.36 109.14  627.0    1.0\n",
        "trend[36]      95.12    0.22   6.34  82.93  90.85  94.96  99.42 107.22  813.0    1.0\n",
        "trend[37]      93.49    0.23   6.86  79.99  88.99  93.39  98.11 107.04  919.0    1.0\n",
        "trend[38]      91.59    0.29   8.35  74.46   86.1  91.49  97.32 107.41  836.0    1.0\n",
        "trend[39]      89.57    0.39  10.73  66.44   83.1  89.71  96.79 109.47  767.0    1.0\n",
        "mm[0]          -8.75    0.33   8.56  -25.8 -14.32  -8.68   -3.2   7.59  656.0    1.0\n",
        "mm[1]         -18.16    0.29   8.45 -34.41 -23.64 -18.16  -12.6  -1.61  827.0    1.0\n",
        "mm[2]         -19.64    0.26   8.14 -36.14 -24.97 -19.63 -14.37  -3.64 1001.0    1.0\n",
        "mm[3]         -26.59     0.4   8.65 -42.52 -32.41 -26.81 -20.94  -8.76  469.0    1.0\n",
        "mm[4]           3.15    0.28    8.4 -14.83  -1.96   3.41   8.82  19.11  921.0    1.0\n",
        "mm[5]          23.95    0.57   9.32   5.39   18.2  23.85  29.63  46.36  270.0   1.01\n",
        "mm[6]          45.04    0.28   8.34   28.7  39.31  44.96  50.57  61.87  871.0    1.0\n",
        "mm[7]          -6.04    0.38   8.56 -21.57  -11.7  -6.41  -0.59  11.78  514.0    1.0\n",
        "mm[8]         -20.19     0.3    7.9 -35.63 -25.49 -20.19 -14.74  -5.01  714.0    1.0\n",
        "mm[9]          -19.5    0.24   7.54 -33.96 -24.52 -19.64 -14.58  -4.51  959.0    1.0\n",
        "mm[10]        -27.67    0.31   7.87 -42.33 -32.91 -27.84 -22.46 -11.43  664.0    1.0\n",
        "mm[11]          5.46    0.25   7.62  -8.98   0.48    5.5  10.56  20.44  948.0    1.0\n",
        "mm[12]         23.82    0.52   8.48    7.1   18.5  23.86  29.09  43.06  263.0   1.01\n",
        "mm[13]         45.09    0.26   7.85  29.93  39.95  45.09  50.02  61.55  887.0    1.0\n",
        "mm[14]          -8.4     0.3    7.5  -22.4 -13.23  -8.48  -3.47    6.9  606.0    1.0\n",
        "mm[15]        -18.74    0.29   7.45 -33.37 -23.85  -18.8 -13.82  -4.26  670.0    1.0\n",
        "mm[16]        -19.41    0.24   7.26 -33.46 -24.38 -19.52  -14.6   -4.6  947.0    1.0\n",
        "mm[17]        -28.75    0.26   7.29 -42.49 -33.88 -28.68 -23.84 -14.72  794.0    1.0\n",
        "mm[18]          6.33    0.25   7.45  -7.98   1.29   6.33  11.01  21.38  869.0    1.0\n",
        "mm[19]         24.39    0.48   8.24   7.42  19.05  24.35  29.54  41.94  290.0   1.01\n",
        "mm[20]         45.57    0.26   7.75   30.4  40.54   45.4  50.63  61.55  888.0    1.0\n",
        "mm[21]        -10.25    0.28   7.58 -25.32 -15.27 -10.09  -5.19    4.8  738.0    1.0\n",
        "mm[22]        -17.56    0.27    7.4 -32.34 -22.54 -17.51 -12.53  -3.07  744.0    1.0\n",
        "mm[23]        -19.82    0.24   7.16 -34.04 -24.53 -19.95 -14.95  -5.69  892.0    1.0\n",
        "mm[24]        -28.41    0.26   7.31 -42.15 -33.74 -28.58 -23.29 -14.35  810.0    1.0\n",
        "mm[25]           5.2    0.24   7.39  -8.99   0.32   5.47  10.13  19.43  952.0    1.0\n",
        "mm[26]         25.42    0.35   8.06  10.24   19.9  25.44   30.6  40.58  536.0   1.01\n",
        "mm[27]         45.62    0.25   7.72  30.59  40.47  45.65  50.68  61.15  931.0    1.0\n",
        "mm[28]        -10.37    0.39    8.0 -26.01 -15.53 -10.21  -4.97   5.02  428.0    1.0\n",
        "mm[29]         -17.0    0.29   7.68 -31.98 -22.36 -17.07 -11.85  -2.07  724.0    1.0\n",
        "mm[30]        -19.98    0.24   7.39 -34.51 -24.91 -20.26 -14.94  -5.79  929.0    1.0\n",
        "mm[31]        -29.84    0.27   7.64 -43.81 -35.27 -29.86 -24.47 -15.35  794.0    1.0\n",
        "mm[32]          5.39    0.26   7.77 -10.19   0.27   5.61  10.81  20.44  864.0    1.0\n",
        "mm[33]         26.93    0.33   8.67  10.73  20.87  26.88  32.85  44.37  712.0    1.0\n",
        "mm[34]         45.52    0.28   8.32  29.51  39.98  45.55  50.64  61.81  897.0    1.0\n",
        "mm[35]        -12.07    0.42    8.7 -29.15 -17.76 -11.77  -6.26    4.8  433.0    1.0\n",
        "mm[36]        -15.79     0.3   8.41 -31.75 -21.59 -15.92 -10.13   1.13  800.0    1.0\n",
        "mm[37]         -19.5    0.26   8.08  -35.6 -24.68 -19.69 -14.12  -3.61  937.0    1.0\n",
        "mm[38]        -30.18    0.29   8.46 -47.17 -35.91 -30.14 -24.33 -14.62  823.0    1.0\n",
        "mm[39]          4.37    0.31   8.82 -14.05  -1.24   4.78  10.67  20.79  801.0    1.0\n",
        "s_trend         1.61    0.08   1.35   0.35    0.7   1.19   2.02   5.36  316.0   1.02\n",
        "s_mm            3.28    0.12   2.34   0.55   1.52    2.7   4.39   9.27  353.0    1.0\n",
        "s_g            16.94    0.06   1.82  13.14  15.69  17.04  18.36  19.77  789.0    1.0\n",
        "mm_all[0]      -8.75    0.33   8.56  -25.8 -14.32  -8.68   -3.2   7.59  656.0    1.0\n",
        "mm_all[1]     -18.16    0.29   8.45 -34.41 -23.64 -18.16  -12.6  -1.61  827.0    1.0\n",
        "mm_all[2]     -19.64    0.26   8.14 -36.14 -24.97 -19.63 -14.37  -3.64 1001.0    1.0\n",
        "mm_all[3]     -26.59     0.4   8.65 -42.52 -32.41 -26.81 -20.94  -8.76  469.0    1.0\n",
        "mm_all[4]       3.15    0.28    8.4 -14.83  -1.96   3.41   8.82  19.11  921.0    1.0\n",
        "mm_all[5]      23.95    0.57   9.32   5.39   18.2  23.85  29.63  46.36  270.0   1.01\n",
        "mm_all[6]      45.04    0.28   8.34   28.7  39.31  44.96  50.57  61.87  871.0    1.0\n",
        "mm_all[7]      -6.04    0.38   8.56 -21.57  -11.7  -6.41  -0.59  11.78  514.0    1.0\n",
        "mm_all[8]     -20.19     0.3    7.9 -35.63 -25.49 -20.19 -14.74  -5.01  714.0    1.0\n",
        "mm_all[9]      -19.5    0.24   7.54 -33.96 -24.52 -19.64 -14.58  -4.51  959.0    1.0\n",
        "mm_all[10]    -27.67    0.31   7.87 -42.33 -32.91 -27.84 -22.46 -11.43  664.0    1.0\n",
        "mm_all[11]      5.46    0.25   7.62  -8.98   0.48    5.5  10.56  20.44  948.0    1.0\n",
        "mm_all[12]     23.82    0.52   8.48    7.1   18.5  23.86  29.09  43.06  263.0   1.01\n",
        "mm_all[13]     45.09    0.26   7.85  29.93  39.95  45.09  50.02  61.55  887.0    1.0\n",
        "mm_all[14]      -8.4     0.3    7.5  -22.4 -13.23  -8.48  -3.47    6.9  606.0    1.0\n",
        "mm_all[15]    -18.74    0.29   7.45 -33.37 -23.85  -18.8 -13.82  -4.26  670.0    1.0\n",
        "mm_all[16]    -19.41    0.24   7.26 -33.46 -24.38 -19.52  -14.6   -4.6  947.0    1.0\n",
        "mm_all[17]    -28.75    0.26   7.29 -42.49 -33.88 -28.68 -23.84 -14.72  794.0    1.0\n",
        "mm_all[18]      6.33    0.25   7.45  -7.98   1.29   6.33  11.01  21.38  869.0    1.0\n",
        "mm_all[19]     24.39    0.48   8.24   7.42  19.05  24.35  29.54  41.94  290.0   1.01\n",
        "mm_all[20]     45.57    0.26   7.75   30.4  40.54   45.4  50.63  61.55  888.0    1.0\n",
        "mm_all[21]    -10.25    0.28   7.58 -25.32 -15.27 -10.09  -5.19    4.8  738.0    1.0\n",
        "mm_all[22]    -17.56    0.27    7.4 -32.34 -22.54 -17.51 -12.53  -3.07  744.0    1.0\n",
        "mm_all[23]    -19.82    0.24   7.16 -34.04 -24.53 -19.95 -14.95  -5.69  892.0    1.0\n",
        "mm_all[24]    -28.41    0.26   7.31 -42.15 -33.74 -28.58 -23.29 -14.35  810.0    1.0\n",
        "mm_all[25]       5.2    0.24   7.39  -8.99   0.32   5.47  10.13  19.43  952.0    1.0\n",
        "mm_all[26]     25.42    0.35   8.06  10.24   19.9  25.44   30.6  40.58  536.0   1.01\n",
        "mm_all[27]     45.62    0.25   7.72  30.59  40.47  45.65  50.68  61.15  931.0    1.0\n",
        "mm_all[28]    -10.37    0.39    8.0 -26.01 -15.53 -10.21  -4.97   5.02  428.0    1.0\n",
        "mm_all[29]     -17.0    0.29   7.68 -31.98 -22.36 -17.07 -11.85  -2.07  724.0    1.0\n",
        "mm_all[30]    -19.98    0.24   7.39 -34.51 -24.91 -20.26 -14.94  -5.79  929.0    1.0\n",
        "mm_all[31]    -29.84    0.27   7.64 -43.81 -35.27 -29.86 -24.47 -15.35  794.0    1.0\n",
        "mm_all[32]      5.39    0.26   7.77 -10.19   0.27   5.61  10.81  20.44  864.0    1.0\n",
        "mm_all[33]     26.93    0.33   8.67  10.73  20.87  26.88  32.85  44.37  712.0    1.0\n",
        "mm_all[34]     45.52    0.28   8.32  29.51  39.98  45.55  50.64  61.81  897.0    1.0\n",
        "mm_all[35]    -12.07    0.42    8.7 -29.15 -17.76 -11.77  -6.26    4.8  433.0    1.0\n",
        "mm_all[36]    -15.79     0.3   8.41 -31.75 -21.59 -15.92 -10.13   1.13  800.0    1.0\n",
        "mm_all[37]     -19.5    0.26   8.08  -35.6 -24.68 -19.69 -14.12  -3.61  937.0    1.0\n",
        "mm_all[38]    -30.18    0.29   8.46 -47.17 -35.91 -30.14 -24.33 -14.62  823.0    1.0\n",
        "mm_all[39]      4.37    0.31   8.82 -14.05  -1.24   4.78  10.67  20.79  801.0    1.0\n",
        "mm_all[40]      4.39    0.32    9.6 -16.38  -1.64   4.99  10.84  22.68  912.0    1.0\n",
        "mm_all[41]      4.32    0.35  10.24  -17.4  -1.96   4.73  11.17   23.1  840.0    1.0\n",
        "mm_all[42]      4.34    0.38  10.88 -19.32  -2.02   4.69  11.28   24.6  826.0    1.0\n",
        "mm_all[43]      4.22    0.38  11.59 -21.02  -2.46   4.87  11.39  25.48  931.0    1.0\n",
        "mm_all[44]      4.29     0.4  12.42 -22.19  -2.37   4.75  11.71  27.64  941.0    1.0\n",
        "mm_all[45]      4.11    0.43  13.21 -25.81  -2.99   4.82  11.74  28.96  941.0    1.0\n",
        "mm_all[46]      4.05    0.45  13.91 -27.25  -3.07   4.93  11.94  30.66  942.0    1.0\n",
        "mm_all[47]      4.21    0.47   14.5 -27.24  -3.14   4.86   12.2  32.17  949.0    1.0\n",
        "mm_all[48]      4.13    0.49  15.05 -28.84  -3.25    4.9  12.49  34.27  933.0    1.0\n",
        "mm_all[49]       4.2     0.5  15.58 -28.56  -3.29   4.84  12.62  34.15  955.0    1.0\n",
        "trend_all[0]   138.2    0.34  10.26 116.49  131.9 138.62 145.23 157.06  900.0    1.0\n",
        "trend_all[1]  136.93    0.27   8.23 119.72 131.74 137.09 142.55 152.52  931.0    1.0\n",
        "trend_all[2]  135.32    0.23   7.07  120.7 130.75 135.41 139.99 149.25  923.0    1.0\n",
        "trend_all[3]  133.44    0.22   6.56 120.22 129.06 133.52  137.5 146.69  871.0    1.0\n",
        "trend_all[4]  131.39    0.23    6.4 119.18 127.29 131.28 135.38 144.51  766.0    1.0\n",
        "trend_all[5]  129.35    0.24   6.42 117.08 125.15 129.13 133.43  142.8  706.0    1.0\n",
        "trend_all[6]  127.06    0.25   6.51 114.97 122.69 126.51 131.06 140.73  677.0    1.0\n",
        "trend_all[7]  124.19    0.24   6.47 112.27 119.88 123.84 127.95 138.79  703.0    1.0\n",
        "trend_all[8]  120.54    0.21   6.02  109.2 116.61 120.37 123.93  133.7  811.0    1.0\n",
        "trend_all[9]  116.62    0.18   5.58 105.99 112.94  116.6 120.18  127.8  917.0    1.0\n",
        "trend_all[10] 112.65    0.21    5.5 101.63 109.04 112.79 116.37 122.93  685.0    1.0\n",
        "trend_all[11] 108.92    0.25   5.75  97.03  105.3 109.13 112.66 119.49  541.0   1.01\n",
        "trend_all[12] 105.43    0.35   6.22  92.34  101.6 105.72 109.49 118.22  315.0   1.01\n",
        "trend_all[13] 102.51     0.4   6.52  88.61  98.52 102.87 106.76 116.34  271.0   1.01\n",
        "trend_all[14] 100.31    0.41   6.53  86.42  96.24 100.54 104.54 113.61  252.0   1.01\n",
        "trend_all[15]  98.81    0.41   6.31  85.76  94.86  98.95 102.83 111.94  239.0   1.01\n",
        "trend_all[16]  97.78    0.39   6.07  85.57  93.79  97.84  101.5 111.53  240.0   1.01\n",
        "trend_all[17]  97.03    0.38   6.05  85.63  92.95  97.05 100.62  112.0  256.0   1.01\n",
        "trend_all[18]  96.35    0.37   6.12  84.94  92.17  96.14  99.91 111.18  276.0   1.01\n",
        "trend_all[19]  95.39    0.37   6.08   84.1  91.24  95.22  99.11 109.93  275.0   1.01\n",
        "trend_all[20]  94.33    0.35   5.96  82.99  90.37  94.28  98.01 108.45  284.0   1.01\n",
        "trend_all[21]   93.4    0.37   5.94  82.02  89.42  93.46   97.1 107.58  255.0   1.01\n",
        "trend_all[22]  92.79    0.38   5.91  81.36  88.92  92.83   96.5 106.61  246.0   1.01\n",
        "trend_all[23]  92.39    0.37   5.98  80.73  88.49  92.45  96.09 106.23  267.0   1.01\n",
        "trend_all[24]  92.06    0.36   5.99  80.29  88.25  92.16  95.75 106.34  278.0   1.01\n",
        "trend_all[25]  91.78    0.35   5.98  80.01  88.03  91.93   95.5  105.9  292.0   1.01\n",
        "trend_all[26]  91.83    0.32   5.92  80.04   88.1  91.73  95.55 104.89  333.0   1.01\n",
        "trend_all[27]  92.21    0.29   5.87  80.34  88.45  92.13  95.95 104.17  410.0   1.01\n",
        "trend_all[28]  92.83    0.25   5.78  81.32  89.02  92.67  96.65 103.62  540.0    1.0\n",
        "trend_all[29]  93.55    0.21   5.73  81.88  89.85   93.4  97.38 103.98  752.0    1.0\n",
        "trend_all[30]  94.47    0.21   5.72  83.65  90.63  94.32  98.24 105.37  767.0    1.0\n",
        "trend_all[31]  95.53    0.22   5.83  84.57  91.49  95.33  99.48 107.41  712.0    1.0\n",
        "trend_all[32]  96.52    0.25   6.16  85.25  92.26  96.28 100.18 109.59  588.0    1.0\n",
        "trend_all[33]  97.06    0.27   6.46  85.95   92.6  96.72 100.76 111.07  552.0   1.01\n",
        "trend_all[34]  96.95    0.27   6.45  85.73  92.55  96.59  101.0 110.45  562.0   1.01\n",
        "trend_all[35]  96.28    0.25   6.37  84.59  91.94  95.84 100.36 109.14  627.0    1.0\n",
        "trend_all[36]  95.12    0.22   6.34  82.93  90.85  94.96  99.42 107.22  813.0    1.0\n",
        "trend_all[37]  93.49    0.23   6.86  79.99  88.99  93.39  98.11 107.04  919.0    1.0\n",
        "trend_all[38]  91.59    0.29   8.35  74.46   86.1  91.49  97.32 107.41  836.0    1.0\n",
        "trend_all[39]  89.57    0.39  10.73  66.44   83.1  89.71  96.79 109.47  767.0    1.0\n",
        "trend_all[40] -274.5    0.86  25.01 -322.5 -291.4 -274.1 -258.2 -222.9  836.0    1.0\n",
        "trend_all[41]   93.4    0.24   7.38  78.61   88.6   93.3  98.15 107.82  946.0    1.0\n",
        "trend_all[42]  91.53    0.31   8.94  72.81  85.94   91.6  97.36 108.12  832.0    1.0\n",
        "trend_all[43]  89.69    0.39   11.0  65.56  83.28  89.88  97.01 109.87  802.0    1.0\n",
        "trend_all[44] -274.6    0.85  25.09 -322.7 -291.6 -274.4 -258.6 -223.1  873.0    1.0\n",
        "trend_all[45]   93.5    0.25   8.04  77.35  88.41  93.23  98.57 109.31 1001.0    1.0\n",
        "trend_all[46]  91.35    0.33   9.45  71.34   85.8  91.29  97.52 108.97  823.0    1.0\n",
        "trend_all[47]  89.82    0.39  11.36  65.05  83.51  90.13  97.16 110.63  844.0    1.0\n",
        "trend_all[48] -274.6    0.86   25.3 -322.6 -291.8 -274.2 -258.7 -222.2  875.0    1.0\n",
        "trend_all[49]  93.43    0.28   8.61  76.06  88.17   93.4  98.89 109.42  977.0    1.0\n",
        "m_next[0]     129.45    0.45   12.1 104.83 121.69 129.86 137.77 151.89  725.0    1.0\n",
        "m_next[1]     118.77    0.36  11.27  96.46 111.65 118.77 126.09 140.61  968.0    1.0\n",
        "m_next[2]     115.68    0.33  10.56  94.93 109.02 115.44 122.77 137.43 1001.0    1.0\n",
        "m_next[3]     106.85    0.51  10.77  86.91   99.6 106.72 114.04 128.41  452.0    1.0\n",
        "m_next[4]     134.54    0.36  10.53 112.92 127.97 134.69 141.41 155.26  879.0    1.0\n",
        "m_next[5]      153.3    0.57  10.99 131.47  145.9 153.21  160.5 175.42  366.0    1.0\n",
        "m_next[6]      172.1    0.37  10.69 151.78 164.95 171.76 179.17  193.5  813.0    1.0\n",
        "m_next[7]     118.15    0.46  10.61  99.72 110.78  117.4  124.8 140.84  535.0    1.0\n",
        "m_next[8]     100.34    0.38   10.1  81.06  93.33 100.36 107.15 120.58  700.0    1.0\n",
        "m_next[9]      97.12     0.3   9.36  78.64  91.01  97.34 103.21 115.62  992.0    1.0\n",
        "y_next[0]     129.45     0.7  21.08  89.22 115.07 129.44 143.42 171.96  900.0    1.0\n",
        "y_next[1]     119.02    0.65  20.48  79.47 105.32 118.93  132.8 159.49 1001.0    1.0\n",
        "y_next[2]      115.5    0.66  20.48  76.08 102.47 115.36 128.92 157.46  952.0    1.0\n",
        "y_next[3]     107.15    0.69  20.62   65.7  93.43 107.85 120.93 147.41  887.0    1.0\n",
        "y_next[4]     134.42    0.63  19.95  96.21 121.13 134.06 147.76 172.82 1001.0    1.0\n",
        "y_next[5]     153.71     0.7  20.33 113.79 140.51 153.49 167.29 195.13  840.0    1.0\n",
        "y_next[6]     172.32    0.66  20.06 134.05 158.49 172.13 185.78 212.07  937.0    1.0\n",
        "y_next[7]     118.39    0.66  20.15  78.75 104.96 118.37 132.04 158.38  925.0    1.0\n",
        "y_next[8]      100.1    0.63  20.04  59.67  86.91 100.18 113.45 139.15 1001.0    1.0\n",
        "y_next[9]      97.29    0.61  19.31  59.96  84.25  97.33  110.7 133.97 1001.0    1.0\n",
        "lp__          -206.3    2.56  36.45 -275.6 -231.0 -206.9 -182.1 -130.9  203.0   1.02\n",
        "\n",
        "Samples were drawn using NUTS(diag_e) at Wed Oct 28 01:49:36 2015.\n",
        "For each parameter, n_eff is a crude measure of effective sample size,\n",
        "and Rhat is the potential scale reduction factor on split chains (at \n",
        "convergence, Rhat=1).\n"
       ]
      },
      {
       "output_type": "stream",
       "stream": "stderr",
       "text": [
        "//anaconda/python.app/Contents/lib/python2.7/multiprocessing/queues.py:390: UserWarning: Pickling fit objects is an experimental feature!\n",
        "The relevant StanModel instance must be pickled along with this fit object.\n",
        "When unpickling the StanModel must be unpickled first.\n",
        "  return send(obj)\n",
        "//anaconda/python.app/Contents/lib/python2.7/multiprocessing/queues.py:390: UserWarning: Pickling fit objects is an experimental feature!\n",
        "The relevant StanModel instance must be pickled along with this fit object.\n",
        "When unpickling the StanModel must be unpickled first.\n",
        "  return send(obj)\n",
        "//anaconda/python.app/Contents/lib/python2.7/multiprocessing/queues.py:390: UserWarning: Pickling fit objects is an experimental feature!\n",
        "The relevant StanModel instance must be pickled along with this fit object.\n",
        "When unpickling the StanModel must be unpickled first.\n",
        "  return send(obj)\n"
       ]
      }
     ],
     "prompt_number": 82
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "plt.plot(ndat,\".-\")\n",
      "plt.plot(res_norm['mm_all'][0][:-10],\".-\")\n",
      "plt.plot(res_norm['trend_all'][0][:-10],\".-\")\n",
      "plt.legend([\"original data\",\"periodic\",\"trend\"])"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "metadata": {},
       "output_type": "pyout",
       "prompt_number": 83,
       "text": [
        "<matplotlib.legend.Legend at 0x1109d2410>"
       ]
      },
      {
       "metadata": {},
       "output_type": "display_data",
       "png": "iVBORw0KGgoAAAANSUhEUgAAAXoAAAEACAYAAAC9Gb03AAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nRU5ErkJggg==\n",
       "text": [
        "<matplotlib.figure.Figure at 0x10fc3d810>"
       ]
      }
     ],
     "prompt_number": 83
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "from scipy.stats.mstats import mquantiles"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [],
     "prompt_number": 100
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "trend_all=zip(*res_norm['trend_all'])"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [],
     "prompt_number": 97
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "print len(trend_all)\n",
      "norm_trend_range_t=[ mquantiles (i) for i in trend_all]\n",
      "norm_trend_range=zip(*norm_trend_range_t)"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": [
        "50\n"
       ]
      }
     ],
     "prompt_number": 110
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "norm_trend_range_t[0]"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "metadata": {},
       "output_type": "pyout",
       "prompt_number": 111,
       "text": [
        "array([ 131.90238198,  138.62298638,  145.22720111])"
       ]
      }
     ],
     "prompt_number": 111
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "r=range(len(norm_trend_range[0][:-10]))\n",
      "fig, a = plt.subplots(1,sharex=True)\n",
      "a.fill_between(r,norm_trend_range[0][:-10],norm_trend_range[2][:-10], color='blue',alpha=0.5)\n",
      "a.plot(norm_trend_range[1][:-10], lw=2, color='black')\n",
      "plt.plot(ndat,\".-\")"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "metadata": {},
       "output_type": "pyout",
       "prompt_number": 121,
       "text": [
        "[<matplotlib.lines.Line2D at 0x110633490>]"
       ]
      },
      {
       "metadata": {},
       "output_type": "display_data",
       "png": "iVBORw0KGgoAAAANSUhEUgAAAXgAAAEACAYAAAC57G0KAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\naN/tW+cN0HEAAAAASUVORK5CYII=\n",
       "text": [
        "<matplotlib.figure.Figure at 0x10f48e490>"
       ]
      }
     ],
     "prompt_number": 121
    },
    {
     "cell_type": "markdown",
     "metadata": {},
     "source": [
      "\u571f\u65e5\u306e\u5f71\u97ff\u3067\u4e0a\u306b\u5f15\u304d\u305a\u3089\u308c\u3066\u3044\u308b"
     ]
    },
    {
     "cell_type": "heading",
     "level": 3,
     "metadata": {},
     "source": [
      "Poisson model\u306e\u8a08\u7b97"
     ]
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "stanmodel_p = pystan.StanModel(model_code=model0)\n",
      "fit = stanmodel_p.sampling(data=dd, iter=10200, warmup=200, thin=10, chains=3, seed=71)\n",
      "print fit\n",
      "res_p=fit.extract()\n",
      "\n",
      "with open(\"model_poisson40\",\"wb\") as fpp:\n",
      "    pkl.dump(res_p,fpp)"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "output_type": "stream",
       "stream": "stderr",
       "text": [
        "INFO:pystan:COMPILING THE C++ CODE FOR MODEL anon_model_64ec0dc1823c9b26110736a4519e0c07 NOW.\n"
       ]
      },
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": [
        "Inference for Stan model: anon_model_64ec0dc1823c9b26110736a4519e0c07.\n",
        "3 chains, each with iter=10200; warmup=200; thin=10; \n",
        "post-warmup draws per chain=1000, total post-warmup draws=3000.\n",
        "\n",
        "                mean se_mean     sd   2.5%    25%    50%    75%  97.5%  n_eff   Rhat\n",
        "trend[0]         4.9  3.4e-3   0.09   4.71   4.84   4.91   4.97   5.08  769.0    1.0\n",
        "trend[1]        4.91  2.2e-3   0.07   4.79   4.87   4.91   4.96   5.04  873.0    1.0\n",
        "trend[2]        4.91  1.8e-3   0.05   4.81   4.88   4.91   4.94   5.02  843.0    1.0\n",
        "trend[3]        4.89  1.5e-3   0.05    4.8   4.87    4.9   4.92   4.99  944.0    1.0\n",
        "trend[4]        4.87  1.8e-3   0.05   4.76   4.85   4.88   4.91   4.96  762.0    1.0\n",
        "trend[5]        4.86  1.6e-3   0.05   4.76   4.83   4.86   4.89   4.95  832.0    1.0\n",
        "trend[6]        4.85  2.3e-3   0.05   4.77   4.82   4.85   4.89   4.95  446.0   1.01\n",
        "trend[7]        4.84  3.1e-3   0.06   4.74    4.8   4.83   4.87   4.96  313.0   1.01\n",
        "trend[8]        4.79  2.5e-3   0.05    4.7   4.76   4.79   4.82    4.9  424.0    1.0\n",
        "trend[9]        4.74  1.9e-3   0.05   4.65   4.71   4.74   4.77   4.84  667.0    1.0\n",
        "trend[10]       4.69  1.6e-3   0.05   4.59   4.66   4.69   4.72   4.78  833.0    1.0\n",
        "trend[11]       4.64  1.7e-3   0.05   4.54   4.61   4.64   4.67   4.73  835.0    1.0\n",
        "trend[12]       4.59  2.0e-3   0.05   4.48   4.56   4.59   4.62   4.68  612.0   1.01\n",
        "trend[13]       4.55  2.5e-3   0.06   4.43   4.51   4.55   4.59   4.65  480.0   1.02\n",
        "trend[14]       4.52  3.6e-3   0.06   4.39   4.49   4.53   4.56   4.62  274.0   1.01\n",
        "trend[15]       4.52  2.1e-3   0.05    4.4   4.49   4.52   4.55   4.61  643.0   1.01\n",
        "trend[16]       4.53  1.6e-3   0.05   4.43   4.49   4.53   4.56   4.62  897.0    1.0\n",
        "trend[17]       4.54  2.0e-3   0.05   4.44    4.5   4.53   4.57   4.64  663.0    1.0\n",
        "trend[18]       4.54  2.8e-3   0.06   4.45    4.5   4.54   4.57   4.68  405.0   1.01\n",
        "trend[19]       4.53  2.3e-3   0.05   4.44    4.5   4.53   4.56   4.65  503.0    1.0\n",
        "trend[20]       4.51  1.9e-3   0.05   4.42   4.48   4.51   4.54   4.61  694.0    1.0\n",
        "trend[21]       4.49  1.9e-3   0.05   4.38   4.46    4.5   4.53   4.59  744.0    1.0\n",
        "trend[22]       4.48  1.8e-3   0.05   4.38   4.45   4.48   4.52   4.58  828.0    1.0\n",
        "trend[23]       4.48  1.7e-3   0.05   4.38   4.45   4.48   4.51   4.58  870.0    1.0\n",
        "trend[24]       4.47  1.7e-3   0.05   4.37   4.44   4.47    4.5   4.57  904.0    1.0\n",
        "trend[25]       4.46  1.6e-3   0.05   4.36   4.44   4.46   4.49   4.55  880.0    1.0\n",
        "trend[26]       4.46  1.6e-3   0.05   4.37   4.43   4.47   4.49   4.56  885.0    1.0\n",
        "trend[27]       4.47  1.6e-3   0.05   4.37   4.44   4.47    4.5   4.57  888.0    1.0\n",
        "trend[28]       4.48  1.6e-3   0.05   4.38   4.45   4.48   4.51   4.57  858.0    1.0\n",
        "trend[29]       4.48  1.9e-3   0.05   4.37   4.45   4.48   4.52   4.57  759.0    1.0\n",
        "trend[30]       4.49  2.1e-3   0.06   4.37   4.46    4.5   4.53   4.59  707.0    1.0\n",
        "trend[31]       4.52  1.8e-3   0.05   4.42   4.48   4.52   4.55   4.61  863.0    1.0\n",
        "trend[32]       4.55  2.1e-3   0.05   4.45   4.51   4.54   4.58   4.66  638.0   1.01\n",
        "trend[33]       4.57  3.2e-3   0.06   4.47   4.53   4.56    4.6   4.69  338.0   1.01\n",
        "trend[34]       4.57  3.0e-3   0.06   4.47   4.53   4.56    4.6   4.69  355.0   1.01\n",
        "trend[35]       4.56  2.3e-3   0.05   4.46   4.52   4.55   4.59   4.67  513.0   1.01\n",
        "trend[36]       4.54  1.9e-3   0.05   4.43    4.5   4.54   4.57   4.64  828.0    1.0\n",
        "trend[37]        4.5  2.0e-3   0.06   4.39   4.47    4.5   4.54   4.62  905.0    1.0\n",
        "trend[38]       4.46  3.3e-3   0.08    4.3   4.41   4.46   4.52   4.62  613.0    1.0\n",
        "trend[39]       4.41  5.5e-3   0.12   4.15   4.33   4.42   4.49   4.63  504.0    1.0\n",
        "mm[0]          -0.12  4.9e-3    0.1  -0.33  -0.18  -0.11  -0.05   0.08  454.0   1.01\n",
        "mm[1]           -0.1  2.8e-3   0.08  -0.25  -0.15   -0.1  -0.04   0.07  905.0    1.0\n",
        "mm[2]          -0.15  2.6e-3   0.08  -0.31  -0.21  -0.15   -0.1-1.3e-3  912.0    1.0\n",
        "mm[3]          -0.18  2.9e-3   0.08  -0.33  -0.23  -0.18  -0.12  -0.02  771.0    1.0\n",
        "mm[4]          -0.07  3.0e-3   0.08  -0.23  -0.13  -0.07  -0.01   0.09  799.0    1.0\n",
        "mm[5]           0.17  3.2e-3   0.08   0.02   0.12   0.17   0.22   0.32  542.0    1.0\n",
        "mm[6]           0.34  2.4e-3   0.07   0.21    0.3   0.34   0.39   0.47  781.0    1.0\n",
        "mm[7]           0.14  3.3e-3   0.09  -0.03   0.09   0.14    0.2   0.31  658.0   1.01\n",
        "mm[8]          -0.28  2.8e-3   0.08  -0.44  -0.33  -0.27  -0.22  -0.12  791.0    1.0\n",
        "mm[9]          -0.18  2.5e-3   0.07  -0.33  -0.23  -0.18  -0.13  -0.04  917.0    1.0\n",
        "mm[10]         -0.25  2.7e-3   0.08   -0.4  -0.31  -0.25   -0.2   -0.1  837.0    1.0\n",
        "mm[11]          0.12  2.6e-3   0.07  -0.02   0.07   0.12   0.17   0.27  783.0    1.0\n",
        "mm[12]          0.18  2.4e-3   0.07   0.05   0.14   0.18   0.23   0.33  850.0    1.0\n",
        "mm[13]          0.37  2.3e-3   0.07   0.23   0.32   0.37   0.41    0.5  836.0    1.0\n",
        "mm[14]         -0.06  2.8e-3   0.08  -0.22  -0.12  -0.06  -0.01   0.08  764.0    1.0\n",
        "mm[15]         -0.21  2.6e-3   0.08  -0.37  -0.26   -0.2  -0.15  -0.05  953.0    1.0\n",
        "mm[16]         -0.16  2.6e-3   0.08  -0.32  -0.22  -0.16  -0.11  -0.01  947.0    1.0\n",
        "mm[17]         -0.32  2.7e-3   0.08  -0.48  -0.37  -0.32  -0.27  -0.16  912.0    1.0\n",
        "mm[18]          0.19  3.0e-3   0.08   0.04   0.13   0.18   0.24   0.35  727.0    1.0\n",
        "mm[19]          0.24  2.7e-3   0.07    0.1   0.19   0.24   0.29   0.38  672.0    1.0\n",
        "mm[20]          0.42  2.2e-3   0.06   0.29   0.37   0.42   0.46   0.54  839.0    1.0\n",
        "mm[21]         -0.17  3.5e-3   0.09  -0.35  -0.23  -0.17  -0.11  -0.01  646.0    1.0\n",
        "mm[22]         -0.16  2.5e-3   0.08  -0.32  -0.21  -0.17  -0.11  -0.02  935.0    1.0\n",
        "mm[23]          -0.2  2.8e-3   0.08  -0.36  -0.25   -0.2  -0.14  -0.04  854.0    1.0\n",
        "mm[24]         -0.25  2.9e-3   0.09  -0.42  -0.31  -0.26   -0.2  -0.08  868.0    1.0\n",
        "mm[25]          0.07  2.5e-3   0.07  -0.07   0.02   0.08   0.12   0.21  868.0    1.0\n",
        "mm[26]          0.26  2.4e-3   0.07   0.11   0.21   0.26   0.31   0.39  889.0    1.0\n",
        "mm[27]          0.44  2.3e-3   0.07    0.3   0.39   0.44   0.48   0.57  916.0    1.0\n",
        "mm[28]         -0.07  3.8e-3   0.08  -0.23  -0.13  -0.07  -0.02   0.11  491.0    1.0\n",
        "mm[29]         -0.18  4.5e-3   0.09  -0.37  -0.23  -0.17  -0.12  -0.03  378.0   1.01\n",
        "mm[30]         -0.24  3.0e-3   0.09  -0.43   -0.3  -0.24  -0.18  -0.08  837.0    1.0\n",
        "mm[31]          -0.4  3.2e-3   0.09  -0.58  -0.45   -0.4  -0.34  -0.23  774.0    1.0\n",
        "mm[32]          0.15  2.4e-3   0.07 2.4e-3    0.1   0.15    0.2    0.3  967.0    1.0\n",
        "mm[33]          0.37  3.2e-3   0.08    0.2   0.31   0.36   0.43   0.53  716.0    1.0\n",
        "mm[34]          0.42  3.1e-3   0.08   0.27   0.37   0.42   0.47   0.56  617.0    1.0\n",
        "mm[35]         -0.21  3.2e-3   0.09  -0.38  -0.27   -0.2  -0.14  -0.03  819.0    1.0\n",
        "mm[36]         -0.08  2.8e-3   0.09  -0.25  -0.13  -0.08  -0.02   0.09  938.0    1.0\n",
        "mm[37]         -0.18  2.9e-3   0.09  -0.35  -0.23  -0.18  -0.12 1.9e-3  957.0    1.0\n",
        "mm[38]         -0.44  3.8e-3   0.11  -0.66  -0.51  -0.44  -0.37  -0.23  787.0    1.0\n",
        "mm[39]          0.04  4.9e-3   0.12  -0.19  -0.03   0.04   0.12   0.28  583.0    1.0\n",
        "s_trend         0.03  1.7e-3   0.03 6.0e-3   0.02   0.02   0.04   0.11  241.0   1.01\n",
        "s_mm            0.11  1.8e-3   0.04   0.04   0.08   0.11   0.13   0.19  450.0   1.01\n",
        "s_g            10.03    0.18   5.66   0.52   5.25  10.02  14.83  19.43  945.0    1.0\n",
        "mm_all[0]      -0.12  4.9e-3    0.1  -0.33  -0.18  -0.11  -0.05   0.08  454.0   1.01\n",
        "mm_all[1]       -0.1  2.8e-3   0.08  -0.25  -0.15   -0.1  -0.04   0.07  905.0    1.0\n",
        "mm_all[2]      -0.15  2.6e-3   0.08  -0.31  -0.21  -0.15   -0.1-1.3e-3  912.0    1.0\n",
        "mm_all[3]      -0.18  2.9e-3   0.08  -0.33  -0.23  -0.18  -0.12  -0.02  771.0    1.0\n",
        "mm_all[4]      -0.07  3.0e-3   0.08  -0.23  -0.13  -0.07  -0.01   0.09  799.0    1.0\n",
        "mm_all[5]       0.17  3.2e-3   0.08   0.02   0.12   0.17   0.22   0.32  542.0    1.0\n",
        "mm_all[6]       0.34  2.4e-3   0.07   0.21    0.3   0.34   0.39   0.47  781.0    1.0\n",
        "mm_all[7]       0.14  3.3e-3   0.09  -0.03   0.09   0.14    0.2   0.31  658.0   1.01\n",
        "mm_all[8]      -0.28  2.8e-3   0.08  -0.44  -0.33  -0.27  -0.22  -0.12  791.0    1.0\n",
        "mm_all[9]      -0.18  2.5e-3   0.07  -0.33  -0.23  -0.18  -0.13  -0.04  917.0    1.0\n",
        "mm_all[10]     -0.25  2.7e-3   0.08   -0.4  -0.31  -0.25   -0.2   -0.1  837.0    1.0\n",
        "mm_all[11]      0.12  2.6e-3   0.07  -0.02   0.07   0.12   0.17   0.27  783.0    1.0\n",
        "mm_all[12]      0.18  2.4e-3   0.07   0.05   0.14   0.18   0.23   0.33  850.0    1.0\n",
        "mm_all[13]      0.37  2.3e-3   0.07   0.23   0.32   0.37   0.41    0.5  836.0    1.0\n",
        "mm_all[14]     -0.06  2.8e-3   0.08  -0.22  -0.12  -0.06  -0.01   0.08  764.0    1.0\n",
        "mm_all[15]     -0.21  2.6e-3   0.08  -0.37  -0.26   -0.2  -0.15  -0.05  953.0    1.0\n",
        "mm_all[16]     -0.16  2.6e-3   0.08  -0.32  -0.22  -0.16  -0.11  -0.01  947.0    1.0\n",
        "mm_all[17]     -0.32  2.7e-3   0.08  -0.48  -0.37  -0.32  -0.27  -0.16  912.0    1.0\n",
        "mm_all[18]      0.19  3.0e-3   0.08   0.04   0.13   0.18   0.24   0.35  727.0    1.0\n",
        "mm_all[19]      0.24  2.7e-3   0.07    0.1   0.19   0.24   0.29   0.38  672.0    1.0\n",
        "mm_all[20]      0.42  2.2e-3   0.06   0.29   0.37   0.42   0.46   0.54  839.0    1.0\n",
        "mm_all[21]     -0.17  3.5e-3   0.09  -0.35  -0.23  -0.17  -0.11  -0.01  646.0    1.0\n",
        "mm_all[22]     -0.16  2.5e-3   0.08  -0.32  -0.21  -0.17  -0.11  -0.02  935.0    1.0\n",
        "mm_all[23]      -0.2  2.8e-3   0.08  -0.36  -0.25   -0.2  -0.14  -0.04  854.0    1.0\n",
        "mm_all[24]     -0.25  2.9e-3   0.09  -0.42  -0.31  -0.26   -0.2  -0.08  868.0    1.0\n",
        "mm_all[25]      0.07  2.5e-3   0.07  -0.07   0.02   0.08   0.12   0.21  868.0    1.0\n",
        "mm_all[26]      0.26  2.4e-3   0.07   0.11   0.21   0.26   0.31   0.39  889.0    1.0\n",
        "mm_all[27]      0.44  2.3e-3   0.07    0.3   0.39   0.44   0.48   0.57  916.0    1.0\n",
        "mm_all[28]     -0.07  3.8e-3   0.08  -0.23  -0.13  -0.07  -0.02   0.11  491.0    1.0\n",
        "mm_all[29]     -0.18  4.5e-3   0.09  -0.37  -0.23  -0.17  -0.12  -0.03  378.0   1.01\n",
        "mm_all[30]     -0.24  3.0e-3   0.09  -0.43   -0.3  -0.24  -0.18  -0.08  837.0    1.0\n",
        "mm_all[31]      -0.4  3.2e-3   0.09  -0.58  -0.45   -0.4  -0.34  -0.23  774.0    1.0\n",
        "mm_all[32]      0.15  2.4e-3   0.07 2.4e-3    0.1   0.15    0.2    0.3  967.0    1.0\n",
        "mm_all[33]      0.37  3.2e-3   0.08    0.2   0.31   0.36   0.43   0.53  716.0    1.0\n",
        "mm_all[34]      0.42  3.1e-3   0.08   0.27   0.37   0.42   0.47   0.56  617.0    1.0\n",
        "mm_all[35]     -0.21  3.2e-3   0.09  -0.38  -0.27   -0.2  -0.14  -0.03  819.0    1.0\n",
        "mm_all[36]     -0.08  2.8e-3   0.09  -0.25  -0.13  -0.08  -0.02   0.09  938.0    1.0\n",
        "mm_all[37]     -0.18  2.9e-3   0.09  -0.35  -0.23  -0.18  -0.12 1.9e-3  957.0    1.0\n",
        "mm_all[38]     -0.44  3.8e-3   0.11  -0.66  -0.51  -0.44  -0.37  -0.23  787.0    1.0\n",
        "mm_all[39]      0.04  4.9e-3   0.12  -0.19  -0.03   0.04   0.12   0.28  583.0    1.0\n",
        "mm_all[40]      0.04  6.1e-3   0.17  -0.33  -0.06   0.05   0.15   0.35  779.0    1.0\n",
        "mm_all[41]      0.04  7.1e-3   0.21   -0.4  -0.08   0.06   0.18   0.42  837.0    1.0\n",
        "mm_all[42]      0.04  8.0e-3   0.23  -0.47   -0.1   0.05   0.19   0.48  856.0    1.0\n",
        "mm_all[43]      0.04  8.9e-3   0.26  -0.54  -0.11   0.06   0.21   0.55  889.0    1.0\n",
        "mm_all[44]      0.04  9.6e-3   0.29  -0.56  -0.12   0.05   0.23   0.61  891.0    1.0\n",
        "mm_all[45]      0.05    0.01   0.31  -0.63  -0.14   0.06   0.24   0.65  944.0    1.0\n",
        "mm_all[46]      0.04    0.01   0.34  -0.67  -0.15   0.06   0.25   0.69  963.0    1.0\n",
        "mm_all[47]      0.05    0.01   0.35   -0.7  -0.16   0.06   0.26   0.73  975.0    1.0\n",
        "mm_all[48]      0.04    0.01   0.37  -0.73  -0.17   0.06   0.27   0.76  962.0    1.0\n",
        "mm_all[49]      0.04    0.01   0.39  -0.77  -0.19   0.06   0.28   0.78  951.0    1.0\n",
        "trend_all[0]     4.9  3.4e-3   0.09   4.71   4.84   4.91   4.97   5.08  769.0    1.0\n",
        "trend_all[1]    4.91  2.2e-3   0.07   4.79   4.87   4.91   4.96   5.04  873.0    1.0\n",
        "trend_all[2]    4.91  1.8e-3   0.05   4.81   4.88   4.91   4.94   5.02  843.0    1.0\n",
        "trend_all[3]    4.89  1.5e-3   0.05    4.8   4.87    4.9   4.92   4.99  944.0    1.0\n",
        "trend_all[4]    4.87  1.8e-3   0.05   4.76   4.85   4.88   4.91   4.96  762.0    1.0\n",
        "trend_all[5]    4.86  1.6e-3   0.05   4.76   4.83   4.86   4.89   4.95  832.0    1.0\n",
        "trend_all[6]    4.85  2.3e-3   0.05   4.77   4.82   4.85   4.89   4.95  446.0   1.01\n",
        "trend_all[7]    4.84  3.1e-3   0.06   4.74    4.8   4.83   4.87   4.96  313.0   1.01\n",
        "trend_all[8]    4.79  2.5e-3   0.05    4.7   4.76   4.79   4.82    4.9  424.0    1.0\n",
        "trend_all[9]    4.74  1.9e-3   0.05   4.65   4.71   4.74   4.77   4.84  667.0    1.0\n",
        "trend_all[10]   4.69  1.6e-3   0.05   4.59   4.66   4.69   4.72   4.78  833.0    1.0\n",
        "trend_all[11]   4.64  1.7e-3   0.05   4.54   4.61   4.64   4.67   4.73  835.0    1.0\n",
        "trend_all[12]   4.59  2.0e-3   0.05   4.48   4.56   4.59   4.62   4.68  612.0   1.01\n",
        "trend_all[13]   4.55  2.5e-3   0.06   4.43   4.51   4.55   4.59   4.65  480.0   1.02\n",
        "trend_all[14]   4.52  3.6e-3   0.06   4.39   4.49   4.53   4.56   4.62  274.0   1.01\n",
        "trend_all[15]   4.52  2.1e-3   0.05    4.4   4.49   4.52   4.55   4.61  643.0   1.01\n",
        "trend_all[16]   4.53  1.6e-3   0.05   4.43   4.49   4.53   4.56   4.62  897.0    1.0\n",
        "trend_all[17]   4.54  2.0e-3   0.05   4.44    4.5   4.53   4.57   4.64  663.0    1.0\n",
        "trend_all[18]   4.54  2.8e-3   0.06   4.45    4.5   4.54   4.57   4.68  405.0   1.01\n",
        "trend_all[19]   4.53  2.3e-3   0.05   4.44    4.5   4.53   4.56   4.65  503.0    1.0\n",
        "trend_all[20]   4.51  1.9e-3   0.05   4.42   4.48   4.51   4.54   4.61  694.0    1.0\n",
        "trend_all[21]   4.49  1.9e-3   0.05   4.38   4.46    4.5   4.53   4.59  744.0    1.0\n",
        "trend_all[22]   4.48  1.8e-3   0.05   4.38   4.45   4.48   4.52   4.58  828.0    1.0\n",
        "trend_all[23]   4.48  1.7e-3   0.05   4.38   4.45   4.48   4.51   4.58  870.0    1.0\n",
        "trend_all[24]   4.47  1.7e-3   0.05   4.37   4.44   4.47    4.5   4.57  904.0    1.0\n",
        "trend_all[25]   4.46  1.6e-3   0.05   4.36   4.44   4.46   4.49   4.55  880.0    1.0\n",
        "trend_all[26]   4.46  1.6e-3   0.05   4.37   4.43   4.47   4.49   4.56  885.0    1.0\n",
        "trend_all[27]   4.47  1.6e-3   0.05   4.37   4.44   4.47    4.5   4.57  888.0    1.0\n",
        "trend_all[28]   4.48  1.6e-3   0.05   4.38   4.45   4.48   4.51   4.57  858.0    1.0\n",
        "trend_all[29]   4.48  1.9e-3   0.05   4.37   4.45   4.48   4.52   4.57  759.0    1.0\n",
        "trend_all[30]   4.49  2.1e-3   0.06   4.37   4.46    4.5   4.53   4.59  707.0    1.0\n",
        "trend_all[31]   4.52  1.8e-3   0.05   4.42   4.48   4.52   4.55   4.61  863.0    1.0\n",
        "trend_all[32]   4.55  2.1e-3   0.05   4.45   4.51   4.54   4.58   4.66  638.0   1.01\n",
        "trend_all[33]   4.57  3.2e-3   0.06   4.47   4.53   4.56    4.6   4.69  338.0   1.01\n",
        "trend_all[34]   4.57  3.0e-3   0.06   4.47   4.53   4.56    4.6   4.69  355.0   1.01\n",
        "trend_all[35]   4.56  2.3e-3   0.05   4.46   4.52   4.55   4.59   4.67  513.0   1.01\n",
        "trend_all[36]   4.54  1.9e-3   0.05   4.43    4.5   4.54   4.57   4.64  828.0    1.0\n",
        "trend_all[37]    4.5  2.0e-3   0.06   4.39   4.47    4.5   4.54   4.62  905.0    1.0\n",
        "trend_all[38]   4.46  3.3e-3   0.08    4.3   4.41   4.46   4.52   4.62  613.0    1.0\n",
        "trend_all[39]   4.41  5.5e-3   0.12   4.15   4.33   4.42   4.49   4.63  504.0    1.0\n",
        "trend_all[40] -13.38    0.01   0.25 -13.85 -13.55 -13.39 -13.21 -12.88  590.0    1.0\n",
        "trend_all[41]    4.5  2.8e-3   0.08   4.34   4.46    4.5   4.55   4.65  871.0    1.0\n",
        "trend_all[42]   4.46  3.6e-3    0.1   4.24    4.4   4.47   4.53   4.64  777.0    1.0\n",
        "trend_all[43]   4.41  6.1e-3   0.13   4.11   4.33   4.43    4.5   4.64  491.0    1.0\n",
        "trend_all[44] -13.37    0.01   0.25 -13.86 -13.54 -13.38  -13.2 -12.86  613.0    1.0\n",
        "trend_all[45]    4.5  3.3e-3    0.1   4.29   4.45   4.51   4.56   4.68  909.0    1.0\n",
        "trend_all[46]   4.46  4.1e-3   0.12    4.2    4.4   4.47   4.53   4.66  859.0    1.0\n",
        "trend_all[47]   4.41  6.4e-3   0.15   4.08   4.33   4.43   4.51   4.65  523.0    1.0\n",
        "trend_all[48] -13.38    0.01   0.26 -13.87 -13.55 -13.39 -13.21 -12.82  613.0    1.0\n",
        "trend_all[49]    4.5  3.7e-3   0.12   4.26   4.45    4.5   4.56   4.71 1001.0    1.0\n",
        "m_next[0]       4.79  2.9e-3   0.09   4.61   4.73   4.79   4.85   4.96  933.0    1.0\n",
        "m_next[1]       4.82  3.2e-3   0.09   4.65   4.76   4.82   4.88   4.99  745.0    1.0\n",
        "m_next[2]       4.76  2.9e-3   0.09   4.58    4.7   4.76   4.82   4.92  940.0    1.0\n",
        "m_next[3]       4.72  3.1e-3   0.09   4.55   4.66   4.72   4.78   4.88  780.0    1.0\n",
        "m_next[4]       4.81  2.8e-3   0.09   4.63   4.75   4.81   4.87   4.97  969.0    1.0\n",
        "m_next[5]       5.03  3.5e-3   0.08   4.87   4.98   5.04   5.09   5.18  499.0    1.0\n",
        "m_next[6]        5.2  2.4e-3   0.07   5.06   5.15    5.2   5.25   5.33  843.0    1.0\n",
        "m_next[7]       4.98  2.8e-3   0.08   4.82   4.92   4.98   5.03   5.13  865.0    1.0\n",
        "m_next[8]       4.52  4.6e-3    0.1   4.32   4.45   4.52   4.58    4.7  469.0   1.01\n",
        "m_next[9]       4.56  2.9e-3   0.09   4.39    4.5   4.56   4.62   4.73  846.0    1.0\n",
        "y_next[0]     120.53     0.5   15.2   92.0  110.0  120.0  130.0  152.0  942.0    1.0\n",
        "y_next[1]      124.3     0.5  15.69   95.0  113.0  124.0  135.0  156.0  969.0    1.0\n",
        "y_next[2]     117.08    0.47   14.8   89.0  107.0  116.0  126.0  148.0 1001.0    1.0\n",
        "y_next[3]     112.25    0.47  14.28   86.0  102.0  112.0  121.0  141.0  937.0    1.0\n",
        "y_next[4]     122.62    0.49  15.26   94.0  112.0  122.0  133.0  154.0  990.0    1.0\n",
        "y_next[5]     154.05    0.66  16.98  122.0  143.0  153.0  165.0  189.0  657.0    1.0\n",
        "y_next[6]     181.38    0.61  18.55  147.0  169.0  181.0  194.0  219.0  930.0    1.0\n",
        "y_next[7]     145.59    0.56  17.21  113.0  134.0  145.0  157.0  181.0  934.0    1.0\n",
        "y_next[8]      91.91    0.47  13.05   68.0   83.0   91.0  100.0  119.0  758.0    1.0\n",
        "y_next[9]      95.99    0.42  12.69   72.0   87.0   96.0  104.0  121.0  914.0    1.0\n",
        "lp__           1.5e4     2.0  21.13  1.5e4  1.5e4  1.5e4  1.5e4  1.5e4  112.0   1.03\n",
        "\n",
        "Samples were drawn using NUTS(diag_e) at Tue Oct 27 08:44:23 2015.\n",
        "For each parameter, n_eff is a crude measure of effective sample size,\n",
        "and Rhat is the potential scale reduction factor on split chains (at \n",
        "convergence, Rhat=1).\n"
       ]
      },
      {
       "output_type": "stream",
       "stream": "stderr",
       "text": [
        "//anaconda/python.app/Contents/lib/python2.7/multiprocessing/queues.py:390: UserWarning: Pickling fit objects is an experimental feature!\n",
        "The relevant StanModel instance must be pickled along with this fit object.\n",
        "When unpickling the StanModel must be unpickled first.\n",
        "  return send(obj)\n",
        "//anaconda/python.app/Contents/lib/python2.7/multiprocessing/queues.py:390: UserWarning: Pickling fit objects is an experimental feature!\n",
        "The relevant StanModel instance must be pickled along with this fit object.\n",
        "When unpickling the StanModel must be unpickled first.\n",
        "  return send(obj)\n",
        "//anaconda/python.app/Contents/lib/python2.7/multiprocessing/queues.py:390: UserWarning: Pickling fit objects is an experimental feature!\n",
        "The relevant StanModel instance must be pickled along with this fit object.\n",
        "When unpickling the StanModel must be unpickled first.\n",
        "  return send(obj)\n"
       ]
      }
     ],
     "prompt_number": 16
    },
    {
     "cell_type": "markdown",
     "metadata": {},
     "source": [
      "Rhat\u3092\u898b\u308b\u3068\u307b\u307c\u53ce\u675f\u3057\u3066\u3044\u308b"
     ]
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "p_trend_all=zip(*res_p['trend_all'])\n",
      "p_trend_range_t=[ mquantiles (i) for i in p_trend_all]\n",
      "p_trend_range=zip(*p_trend_range_t)"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [],
     "prompt_number": 122
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "pe_trend_range=[np.exp(i) for i in p_trend_range]"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [],
     "prompt_number": 124
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "fig, a = plt.subplots(1,sharex=True)\n",
      "a.fill_between(range(len(pe_trend_range[0][:-10])),pe_trend_range[0][:-10],pe_trend_range[2][:-10], color='blue',alpha=0.5)\n",
      "a.plot(pe_trend_range[1][:-10], lw=2, color='black')\n",
      "plt.plot(ndat,\".-\")"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "metadata": {},
       "output_type": "pyout",
       "prompt_number": 125,
       "text": [
        "[<matplotlib.lines.Line2D at 0x112c85cd0>]"
       ]
      },
      {
       "metadata": {},
       "output_type": "display_data",
       "png": "iVBORw0KGgoAAAANSUhEUgAAAXgAAAEACAYAAAC57G0KAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\noiiKoiiKoijBy/8H5ywQv5swoIsAAAAASUVORK5CYII=\n",
       "text": [
        "<matplotlib.figure.Figure at 0x110623110>"
       ]
      }
     ],
     "prompt_number": 125
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "p_mm_all=zip(*res_p['mm_all'])\n",
      "p_mm_range=zip(*[ mquantiles (i) for i in p_mm_all])"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [],
     "prompt_number": 126
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "pe_mm_range=[np.exp(i) for i in p_mm_range]"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [],
     "prompt_number": 127
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "fig, a = plt.subplots(1,sharex=True)\n",
      "a.fill_between(range(len(pe_mm_range[0][:-10])),pe_mm_range[0][:-10],pe_mm_range[2][:-10], color='blue',alpha=0.5)\n",
      "a.plot(pe_mm_range[1][:-10], lw=2, color='black')\n",
      "#plt.plot(ndat,\".-\")"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "metadata": {},
       "output_type": "pyout",
       "prompt_number": 129,
       "text": [
        "[<matplotlib.lines.Line2D at 0x1181a9cd0>]"
       ]
      },
      {
       "metadata": {},
       "output_type": "display_data",
       "png": "iVBORw0KGgoAAAANSUhEUgAAAXUAAAEACAYAAABMEua6AAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nAAAASUVORK5CYII=\n",
       "text": [
        "<matplotlib.figure.Figure at 0x112c70a90>"
       ]
      }
     ],
     "prompt_number": 129
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "plt.plot(res_p['mm_all'][0][:-10],\".-\")\n",
      "plt.plot(res_p['trend_all'][0][:-10],\".-\")\n",
      "plt.legend([\"periodic\",\"trend\"])"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "metadata": {},
       "output_type": "pyout",
       "prompt_number": 39,
       "text": [
        "<matplotlib.legend.Legend at 0x110042a10>"
       ]
      },
      {
       "metadata": {},
       "output_type": "display_data",
       "png": "iVBORw0KGgoAAAANSUhEUgAAAXMAAAEACAYAAABBDJb9AAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nA8QJEPs3qQ2vAAAAAElFTkSuQmCC\n",
       "text": [
        "<matplotlib.figure.Figure at 0x10f80f050>"
       ]
      }
     ],
     "prompt_number": 39
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "fig, a = plt.subplots(1,sharex=True)\n",
      "\n",
      "a.fill_between(range(len(p_trend_range[0][:-10])),p_trend_range[0][:-10],p_trend_range[2][:-10], color='blue',alpha=0.5)\n",
      "a.plot(p_trend_range[1][:-10], lw=2, color='black')\n",
      "\n",
      "a.fill_between(range(len(p_mm_range[0][:-10])),p_mm_range[0][:-10],p_mm_range[2][:-10], color='red',alpha=0.5)\n",
      "a.plot(p_mm_range[1][:-10], lw=2, color='black')"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "metadata": {},
       "output_type": "pyout",
       "prompt_number": 131,
       "text": [
        "[<matplotlib.lines.Line2D at 0x112d45650>]"
       ]
      },
      {
       "metadata": {},
       "output_type": "display_data",
       "png": "iVBORw0KGgoAAAANSUhEUgAAAXMAAAEACAYAAABBDJb9AAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n",
       "text": [
        "<matplotlib.figure.Figure at 0x1181b1510>"
       ]
      }
     ],
     "prompt_number": 131
    },
    {
     "cell_type": "markdown",
     "metadata": {},
     "source": [
      "\u30c8\u30ec\u30f3\u30c9\u306e\u5909\u52d5\u304c\u5927\u304d\u3044"
     ]
    },
    {
     "cell_type": "heading",
     "level": 3,
     "metadata": {},
     "source": [
      "\u6bd4\u8f03"
     ]
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "res_norm=pkl.load(open(\"model_normal40\"))"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [],
     "prompt_number": 31
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "fig, a = plt.subplots(1,sharex=True,figsize=(10,10))\n",
      "\n",
      "a.fill_between(range(len(norm_trend_range[0][:-10])),norm_trend_range[0][:-10],norm_trend_range[2][:-10], color='blue',alpha=0.5)\n",
      "a.plot(norm_trend_range[1][:-10], lw=2, color='darkgray')\n",
      "\n",
      "a.fill_between(range(len(pe_trend_range[0][:-10])),pe_trend_range[0][:-10],pe_trend_range[2][:-10], color='green',alpha=0.5)\n",
      "a.plot(pe_trend_range[1][:-10], lw=2, color='black')\n",
      "plt.plot(ndat,\".-\")\n",
      "plt.legend([\"trend normal model\",\"trend poisson model\"])"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "metadata": {},
       "output_type": "pyout",
       "prompt_number": 144,
       "text": [
        "<matplotlib.legend.Legend at 0x1184b4610>"
       ]
      },
      {
       "metadata": {},
       "output_type": "display_data",
       "png": "iVBORw0KGgoAAAANSUhEUgAAAlcAAAJPCAYAAABRvvFyAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n",
       "text": [
        "<matplotlib.figure.Figure at 0x10f8f8f10>"
       ]
      }
     ],
     "prompt_number": 144
    },
    {
     "cell_type": "markdown",
     "metadata": {},
     "source": [
      "1\u672c\u3060\u3051\u30d7\u30ed\u30c3\u30c8"
     ]
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "plt.plot(res_norm['trend_all'][0][:-10],\".-\")\n",
      "plt.plot(np.exp(res_p['trend_all'][0][:-10]),\".-\")\n",
      "plt.legend([\"trend normal model\",\"trend poisson model\"])"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "metadata": {},
       "output_type": "pyout",
       "prompt_number": 33,
       "text": [
        "<matplotlib.legend.Legend at 0x10fc79c10>"
       ]
      },
      {
       "metadata": {},
       "output_type": "display_data",
       "png": "iVBORw0KGgoAAAANSUhEUgAAAXgAAAEACAYAAAC57G0KAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nYII=\n",
       "text": [
        "<matplotlib.figure.Figure at 0x10f5bb2d0>"
       ]
      }
     ],
     "prompt_number": 33
    },
    {
     "cell_type": "heading",
     "level": 2,
     "metadata": {},
     "source": [
      "Reference"
     ]
    },
    {
     "cell_type": "markdown",
     "metadata": {},
     "source": [
      "- poisson distribution timie series model http://ito-hi.blog.so-net.ne.jp/2014-01-24\n",
      "- http://xiangze.hatenablog.com/entry/2014/03/16/155826\n",
      "- https://gist.github.com/xiangze/9451645#file-model1-stan\n",
      "- https://github.com/iwanami-datascience/vol1/blob/master/matsuura/example2/example2.ipynb\n",
      "- https://github.com/iwanami-datascience/vol1/blob/master/matsuura/example2/model.stan\n",
      "- prior distributions http://heartruptcy.blog.fc2.com/blog-entry-140.html"
     ]
    }
   ],
   "metadata": {}
  }
 ]
}

kancolle_201510.csv

124
82
107
118
155
93
80
79
92
122
142
127
81
89
71
66
74
108
111
97
85
79
142
90
98
134
91
195
580
461
341
358
312
320
265
212
178
147
171
178
112
131
121
113
112
150
184
167
89
93
85
121
115
130
78
72
80
68
125
125
139
66
75
76
75
85
106
138
91
69
61
57
119
147
148
75
92
80
53
81
114
112
70
49
76
65
115
126
126
70
86
86
64
75
99
106
80
63
63
74
80
86
126
57
60
53
62
187
142
134
141
60
68
76
68
108
107
72
68
66
74
99
117
113
71
75
80
69
81
97
131
126
367
281
299
336
263
278
215
181
205
177
174
250
237
148
166
139
150
136
164
185
118
120
117
119
110
135
216
210
119
98
101
85
109
152
78
76
77
79
82
94
119
104
118
125
91
134
99
116
63
61
80
81
89
94
117
49
68
59
63
109
115
103
125
56
67
64
99
78
12