nb/fred-employ-nfp.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Employment data: NFP\n",
    "\n",
    "We examine U.S. employment data, in particular, \n",
    "**Nonfarm Payroll**, which has considerable influence \n",
    "on the financial markets. \n",
    "\n",
    "We focus on forecasting the monthly change in NFP\n",
    "using a variety of optics:\n",
    "\n",
    "- Baseline expectation since 1939\n",
    "- Holt-Winters method\n",
    "- Visual selection of local range\n",
    "- Regression against S&P 500\n",
    "\n",
    "The ideal point forecast would a weighted combination of\n",
    "different techniques, but the standard errors\n",
    "are inherently very large due to survey measurement error.\n",
    "\n",
    "Examining the NFP data, much more volatility is involved than\n",
    "what is commonly understood in the financial news.\n",
    "Secondarily, we are left to also consider the\n",
    "difference from \"street\" expectations\n",
    "to assess possible impact on the markets\n",
    "and the Fed's monetary policy."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "*Dependencies:*\n",
    "\n",
    "- Repository: https://github.com/rsvp/fecon235\n",
    "- Python: matplotlib, pandas\n",
    "     \n",
    "*CHANGE LOG*\n",
    "\n",
    "    2016-05-24  Consider SPX and lagged corporate earnings.\n",
    "    2015-11-12  Preliminary draft."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "from fecon235.fecon235 import *"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      " ::  Python 2.7.11\n",
      " ::  IPython 4.2.0\n",
      " ::  jupyter_core 4.1.0\n",
      " ::  notebook 4.1.0\n",
      " ::  matplotlib 1.5.1\n",
      " ::  numpy 1.10.4\n",
      " ::  pandas 0.18.0\n",
      " ::  pandas_datareader 0.2.1\n",
      " ::  Repository: fecon235 v4.16.0329 devNFP\n",
      " ::  Timestamp: 2016-05-25, 23:03:04 UTC\n",
      " ::  $pwd: /media/yaya/virt15h/virt/dbx/Dropbox/ipy/fecon235/nb\n"
     ]
    }
   ],
   "source": [
    "#  PREAMBLE-p6.16.0428 :: Settings and system details\n",
    "from __future__ import absolute_import, print_function\n",
    "system.specs()\n",
    "pwd = system.getpwd()   # present working directory as variable.\n",
    "print(\" ::  $pwd:\", pwd)\n",
    "#  If a module is modified, automatically reload it:\n",
    "%load_ext autoreload\n",
    "%autoreload 2\n",
    "#       Use 0 to disable this feature.\n",
    "\n",
    "#  Notebook DISPLAY options:\n",
    "#      Represent pandas DataFrames as text; not HTML representation:\n",
    "import pandas as pd\n",
    "pd.set_option( 'display.notebook_repr_html', False )\n",
    "from IPython.display import HTML # useful for snippets\n",
    "#  e.g. HTML('<iframe src=http://en.mobile.wikipedia.org/?useformat=mobile width=700 height=350></iframe>')\n",
    "from IPython.display import Image \n",
    "#  e.g. Image(filename='holt-winters-equations.png', embed=True) # url= also works\n",
    "from IPython.display import YouTubeVideo\n",
    "#  e.g. YouTubeVideo('1j_HxD4iLn8', start='43', width=600, height=400)\n",
    "from IPython.core import page\n",
    "get_ipython().set_hook('show_in_pager', page.as_hook(page.display_page), 0)\n",
    "#  Or equivalently in config file: \"InteractiveShell.display_page = True\", \n",
    "#  which will display results in secondary notebook pager frame in a cell.\n",
    "\n",
    "#  Generate PLOTS inside notebook, \"inline\" generates static png:\n",
    "%matplotlib inline   \n",
    "#          \"notebook\" argument allows interactive zoom and resize."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "collapsed": true
   },
   "source": [
    "## NFP data\n",
    "\n",
    "Federal Reserve FRED states: \"Total **Nonfarm Payroll** is a measure of the number of \n",
    "U.S. workers in the economy that *excludes proprietors, private household \n",
    "employees, unpaid volunteers, farm employees, and the unincorporated self-employed*. \n",
    "This measure accounts for approximately 80 percent of the workers who \n",
    "contribute to Gross Domestic Product (GDP).\"\n",
    "\n",
    "It provides useful insights into the current economic situation \n",
    "because it is a good proxy for the number of jobs added or lost in an economy. \n",
    "Increases in employment might indicate that businesses are hiring which \n",
    "might also suggest that businesses are growing. Additionally, those who \n",
    "are newly employed have increased their personal incomes, which \n",
    "means (all else constant) their disposable incomes have also increased, \n",
    "thus fostering further economic expansion.\n",
    "\n",
    "Generally, the U.S. labor force and levels of unemployment \n",
    "are subject to fluctuations due to seasonal changes in weather, major holidays, \n",
    "and the opening and closing of schools. The Bureau of Labor Statistics (BLS) \n",
    "adjusts the data to offset the seasonal effects to show non-seasonal changes: \n",
    "for example, women's participation in the labor force; or a general decline \n",
    "in the number of employees, a possible indication of a downturn in the economy. \n",
    "\n",
    "To closely examine seasonal and non-seasonal changes, the BLS releases \n",
    "two **monthly** statistical measures: the seasonally adjusted All Employees: \n",
    "Total Nonfarm (PAYEMS) and All Employees: Total Nonfarm (PAYNSA), \n",
    "which is not seasonally adjusted.\"\n",
    "\n",
    "**The financial market keenly watches PAYEMS, thus we set *m4nfp* to that measure.**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "#  Nonfarm Payroll workers in thousands, seasonally adjusted:\n",
    "nfp = get( m4nfp )"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "#  We know the number of total workers, \n",
    "#  so test the assertion that NFP\n",
    "#  \"accounts for approximately 80 percent of the workers.\"\n",
    "workers = get( m4workers )\n",
    "nfp_ratio = todf( nfp / workers )"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "# # Uncomment to see plot:\n",
    "# plot( nfp_ratio )\n",
    "\n",
    "#  nfp_ratio gradually rose from 55% in 1950's\n",
    "#            to about 75% currently."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "So NFP currently accounts for **approximately 75% \n",
    "of all workers** *currently*. The nfp_ratio plot \n",
    "clearly shows the shift away from farms and into \n",
    "corporations since WWII."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAYcAAAEaCAYAAAD65pvjAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nAABJRU5ErkJggg==\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0xaafae04c>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "#  MAIN PLOT of plain NFP, in thousands of workers,\n",
    "#            seasonally adjusted monthly.\n",
    "plot( nfp )"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      " ::  regresstime slope = 128.909629362\n"
     ]
    }
   ],
   "source": [
    "#  NFP has a linear trend over the long-term...\n",
    "nfp_trend = trend( nfp )\n",
    "\n",
    "#  ... dataframe contains points of the computed uptrend."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "#  Set slope for later calculation:\n",
    "nfp_slope = 128.91\n",
    "#  as of 2016-05-20: 128.91\n",
    "#  as of 2016-04-19: 128.92\n",
    "#  as of 2015-11-12: 128.98"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "The slope of the trend gives us a **baseline expectation: \n",
    "on average, 129,000 nonfarm workers can be expected to be added monthly** \n",
    "(given data from 1939 to 2016). We suspect that at the core, \n",
    "this trend could be modeled mainly by population dynamics,\n",
    "but it is also related to the birth and death rate of business firms."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "#  Take the first difference, i.e. Month over Month:\n",
    "nfp_dif = dif( nfp, 1 )"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "#  Subtract the long-term base expected change:\n",
    "nfp_dif_base = todf( nfp_dif - nfp_slope )"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAYMAAAEaCAYAAADzDTuZAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nAABJRU5ErkJggg==\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0xa97d108c>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "#  Our data goes back to 1939, \n",
    "#  so let's plot a recent subsegment:\n",
    "plot( nfp_dif_base['2000':] )"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "The last plot depicts deviations from nfp_slope,\n",
    "that is, deviations from our baseline expectation.\n",
    "\n",
    "During the Great Recession, there were several months where about \n",
    "one million jobs were lost each month.\n",
    "Consistent above average job additions started after 2011. \n",
    "\n",
    "The cumulative effect can be seen by looking at the deviations from trend."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "nfp_dev = nfp - nfp_trend"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAYkAAAEaCAYAAADkL6tQAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz\nAAALEgAACxIB0t1+/AAAIABJREFUeJztnXe4FdXV/z+LZqEIiHQUUBCsYMEWFTXWWNAkakwEjMb4\nYooxNhJ/MUVfNUUxyYtGxYJJxE7U2CNXBUGwAFIUFJEiIqEpRRTu+v2x5njPvZzeZuae9Xme89xz\n9uzZ8z1z58yavdbea4uq4jiO4zipaBK2AMdxHCe6uJFwHMdx0uJGwnEcx0mLGwnHcRwnLW4kHMdx\nnLS4kXAcx3HS4kbCCR0RqRWR3mHrcBxna9xIVCEi8oGIHF3hY3YRkcVpNhc9WUdEJojI9zNs7yMi\n40XkExH5r4g8LSJ9G9T5mYgsE5E1InKniDRP2naxiEwTkc9F5K4U7W8nIqNFZIWIrBaRmix6jxGR\nuSKyTkT+IyI7p6jTPKizKIfvf2PwvVaIyA0Ntv1WRGaKyJci8qsc2sp0HtqJyGOB7g9E5DvFfM9M\nulO0tYuIvCgi60Vkjogc02D7OSKyUEQ+E5FHRaRttu/qZMeNhFMpTgKeTrNNKnD8tsC/gL5AJ2Ba\n8NkEiBwPXAEcBewC7Ar8Jmn/pcDvgDFp2r8jOMbuQHvgZ+mEiMiOwCPAL4O6bwAPpKh6BbA82xcT\nkR8CpwJ7A/sAp4jIhUlV5gOXA0/m0Fa28zAa+BzYCfgecKuI9E/TVsbvmYPuhtwftNEeuBp4ODgG\nIrIncBvwXez/uxG4Ndv3dXJAVf1VRS9gLLAFWA98ClyG3QxqgeHAImAl8EPgAGAGsAr4S1Ibw4CJ\nwF+ANcAc4Ogsx30EGJJmWy3QO3h/EvAmsBb4ELgmqd42wH3Af4HVwGvYzepaYDOwIfhOf87hPLQL\njtsu+PwP4Nqk7UcBy1Ls9zvgrgZluwfnoVWO/4MfABOTPm8faO+bVNYLmA0cDyzK0t4k4IKkz+cB\nr6aodx/wqyxtpT0Pgc5NwK5J2+8F/reQ75mr7mBbH+zG3zKp7CXgwuD9dcDfk7b1DrS2zPR9/ZX9\n5T2JKkNVh2KG4GRVbaOqf0zaPAjYDTgLGAX8Ajga2As4U0QOT6p7EPaEuiPwayBt915EmgFHAM/n\nIHEdcK6q7gB8A7hIRE4Ntg0D2gDdsKfJi4CNqno18Arwo+A7/SSH4xyJ3fxWB5/3xAxighlARxFp\nl0NbgzCD9tvAbTJDRM7IUL/esVR1A/BeUJ7gz8BI7Kk9G6m075mmbiFtJc5DX+BLVX0/3bECV9uh\nqdpK8T0z6haRJ0TkiqS6C1R1fZr6DY+1ADMS9VyKTv64kaheGrp4FPitqn6hqi9gPY37VXWlqn6E\n3YQHJtVfrqp/VtUtqvog8C52U0/FEcD0Bj/wlKjqy6o6O3g/CxiH3dABvsSMUl813lLVdbl93TpE\npDvwV+q7hFphvZcEn2LnqHUOTXbHXCargS7Aj4F7RWT3NPUbHitxvNaBvtOBJqr6eA7HTqe9VY77\n5tJW4jy0Cj7TYPtX50hV26nqq2naalg/o25VPUVVf19gW1tpcwrDjYSTzCdJ7zdS3x++kfo3nqUN\n9v0Q6Jqm3ZOAp3IRICIHBcHJT0RkDeb26hBsvg94FhgnIkuCoGfTXNpNan+noI2/BsYtwTqsl5Jg\nB8xwfpZDsxuBLzA3zWZVfRmYABwnIj2CQOpnIpK4wTY8VuJ4n4nI9sCNQKI3VM+Yi8jIRFsiMjqD\n9ryNZ4a2Euchre4c22pYPx/d+baVTZuTI24kqpNSpP7t1uDzzsBHaermbCQwn/h4oJuqtgX+RnCj\nDG7Av1PVPYFDgZOBocF+Wb9T4A57Fhivqg1H0swG9k36PADrLa0mOzMTh0gq00DzYlVtHbwSN7HZ\nQfsJXS2xAPFszPe+C/CKiCzDYjldReQjEdlZVa9PtKWqIzJon52D7lRkOg/zgGYismvS9n0zHCvd\n95xVgO7ZQO+gjVTHrtdWoLF5oNkpAjcS1cnHWGAvmXxHGHUUkR+LSDMR+TbQjxSGQER6Ai1U9d0c\n220FrFbVL0VkEHBOUluDRWQvEWmCPTl+iQXhwXo9aedaiEhr4DkskPrLFFXGAueLSP/A/341cHfS\n/k1FZFugKXaj3CapF/MyFucZGdQ7DBiMGaRUPAbsKSKni8g2wDWYO24edgPtgd0w9wUuwP5f+wLp\nhhCPBS4Vka4i0g24tIH2ZoH2JkDzQHu6337a8xDEFB7FYi/bi8jXgFOwHl4+33N+LrqTCfaZDlwT\n6D8Di5U9ElT5BzY66rDAkPwWeCQXF6eThbAj5/6q/AsbdvghNmrpUuzJdQvmB0/UWQQckfR5LPCL\n4P0wLEbxZ2xUzzvAMWmOdTFZRhsFx06MbjoDWIj5lx8PjjE22HZ2cKzPgGXAzQnNwMFYXGQlMCrF\nMYYGx/ks6fUp0D2pziXYDXkNcCfQPGnbNdhoqC1Jr18lbe8PvBq0Ows4Nct3PhqYi8V+XgR2TlPv\nSLKMbgrq3RB89/8C1zfYdncK7UMztJXpPLTDbv7rgv/TWQ32/Qw4LNfvmUX3U8BVSZ93xtx4G4I2\nj2pQ/2zsuv4MM2Ztw/6tNYaXBCe3KERkDNb1X66q+wRl12BD4BJ+7l+o6jPBtpHA97Fhiz9V1eeC\n8v2Ae4BtgadU9ZKgvAV2k9o/uJjOUtWsE4yc8iAiw4DzVfWIHOr+Gxs++0z5lTmOU2pK5W66GxvP\n3ZCbVHW/4JUwEP2BM7EnrxOB0SKScHXcit18+gJ9g4k9AOcDq1S1DzY08/c4cWFC8HIcJ4aUxEio\n6kRs+F9DUvm5TwPGqQUhF2Jj7QeJSGegtapOC+qNBYYk7XNv8P5hoN50fCe6qOofVXVT2DocxymM\ncgeufyQi08Xyv+wQlHWjfgBuaVDWDViSVL6EuhE0X+2jqluANSLSvqzKnbSo6r25uJocx4k/5TQS\no7Fg5AAsCPanErZdiVw/juM4VU+zcjWsqiuSPt4BPBG8X4oN8UvQPShLV568z0fBsMM2qrqq4TFF\npBTj/x3HcaoOVU358F3KnoSQ9IQfxBgSnEHdBJrHgbNFpIWI9MJyBU1V1Y+BtSIyKAhkD6UuS+fj\n2LBLgG9jQ+lSkm041zXXXBP6kLJcX3HSGje9rtX1xk1rOfVmoiQ9CRH5JzZ5aEex3PfXAEeJyABs\nfPZCLL0CqjpHRB7EMod+CYzQOpUXU38IbGLY5BjgPhGZj42pPrtQrQsXLix014oTJ60QL72utXzE\nSW+ctEI4ektiJFT1nBTFKWdOBvWvB65PUf4GliitYfkmbNis4ziOU0GqLi3H8OHDw5aQM3HSCvHS\n61rLR5z0xkkrhKO3JDOuo4KIaGP6Po7jOJVARNAKBK5jQU1NTdgSciZOWiFeel1r+YiT3jhphXD0\nVp2RcBzHcXLH3U2O4zhVjrubHMdxnIKoOiMRJx9knLRCvPS61vIRJ71x0goek3Acx3EihsckHMdx\nqhyPSTiO4zgFUXVGIk4+yDhphXjpda3lI05646QVPCbhOI7jRAyPSTiO41Q5HpNwHMdxCqLqjESc\nfJBx0grx0utay0ec9MZJK3hMwnEcx4kYHpNwHMepcjwm4TiO4xRE1RmJOPkg46QV4qXXtZaPOOmN\nk1bwmITjOI4TMTwm4TiOU+V4TMJxHMcpiKozEnHyQcZJK8RLr2stH3HSGyet4DEJx3EcJ2J4TMJx\nHKfK8ZiE4ziOUxBVZyTi5IOMk1aIl17XWj7ipDdOWsFjEo7jOE7E8JiE4zhOleMxCcdxHKcgqs5I\nxMkHGSetEC+9rrV8xElvnLSCxyQcx3GciOExCcdxnCrHYxKO4zjAhg3w+uthq4gXVWck4uSDjJNW\niJde11o+oqz36qvhwANhzRr7HGWtqfCYhOM4ThmZNs3+nn56uDriREliEiIyBjgZWK6q+wRl7YAH\ngF2AhcCZqro22DYS+D6wGfipqj4XlO8H3ANsCzylqpcE5S2AscD+wH+Bs1R1UQodHpNwHCcltbWw\n/fbw4otw2GH2/uWXYf/9w1YWPpWISdwNHN+g7CrgBVXdHXgRGBmI2QM4E+gPnAiMFpGEuFuB81W1\nL9BXRBJtng+sUtU+wCjg9yXS7ThOlfDxx9CmDRx6KAwfbvGJBx4IW1X0KYmRUNWJwOoGxacB9wbv\n7wWGBO9PBcap6mZVXQjMBwaJSGegtaoGHULGJu2T3NbDwDGFao2TDzJOWiFeel1r+Yiq3vfeg65d\n7f0dd8Cbb8Jtt9Wwbl24uvKhscUkOqrqcgBV/RjoGJR3AxYn1VsalHUDliSVLwnK6u2jqluANSLS\nvnzSHcdpbFx6Key3n71v1gwGDoSePeHVV0OVFXkqGbguZbAgpe8sFwYPHlxCGeUlTlohXnpda/mI\not4lS2DhQvjb3+qXH374YN59NxRJBRHGuW1WxraXi0gnVV0euJI+CcqXAj2S6nUPytKVJ+/zkYg0\nBdqo6qpUBx0+fDg9e/YEoG3btgwYMOCrE5voqvln/+yfq+vzSy9B//41TJpUf/vnn8OyZeHrq/Tn\nmpoa7rnnHoCv7pdpUdWSvICewNtJn28ErgzeXwncELzfA3gLaAH0At6jbpTVFGAQ1lN4CjghKB8B\njA7en43FNFJp0GxMmDAha52oECetqvHS61rLRxT1/uAHqqNGbV1++eUTdNiwisspmHKd2+DemfLe\nXhJ3k4j8E3gVG5G0SETOA24AjhWRd7FA8w3BXXwO8CAwJzAEIwKRABcDY4B5wHxVfSYoHwN0EJH5\nwCXYyKmyoAqzZ5erdcdxwmDaNDjkkK3LO3SARVsNpneS8dxNDbjvPhg6FB591CfcOE5jQBVat4al\nS2GHHepvW7kSeve2v83K6XyPOJ67KQ/GjIEDDoAzzoBPPw1bjeM4xfL++2YkGhoIgB13hJ12gvnz\nK68rLlSdkUgEb1Lx4Ycwaxbcf799fvPNymhKRyatUSROel1r+Yia3h/9CL7//dTbampq6N+f2Ixw\nCuPcVp2RyMTll8P3vge77QYjRsCMGWErchynGNavt9Qbv/pV+jr9+sE771ROU9zwmERAwje5bJnl\ndPnzn+3p4v/+r8QiHcepGC++CCNHwmuvpa9z550waRLcfXfldEUNj0nkwNNPw9FHm4EA6NPH/ZSO\nE3cefhhOPjlzHe9JZKbqjEQ6n96UKXDkkXWf99oLnn8eHn+8MrpSETXfbjbipNe1lo+o6K2thb//\nHS64IH2dmpqar4xEHJwqHpMIkcWLYZdd6j53725/TzsNvvwyHE2O4xTOvHmW9bVLl8z1OnSApk3h\nk08y16tWPCYRMHCg+SaTc8svXw5f+xo89pj1LBzHiQ/77GMxxhUrstc96CAYNSr1hLtqwGMSObB4\nMey8c/2yTp1gwAAf5eQ4caRpU3jyydzqduliBsXZmqozEql8ehs22FC5Dh22rr/vvuEZiaj4dnMl\nTnpda/mIgt6334bp0y0onYmE1rgYCY9JhMTixdCjB0iKzpb3JBwnfjz8MJxwQupZ1qno0gU++qi8\nmuKKxySwPE133GHDYBuyeDEceKAtfeg4TjwYMgTOOQfOPDO3+nfeaYsP3XVXeXVFFY9JZOHpp+HE\nE1Nv694d1q4lVkscOk41s2ULvPIKHHZY7vvExd0UBlVnJFL59N59F/beO3V9ETMUS5em3l5OouDb\nzYc46XWt5SNsvY8/DrvuCt26Za/rMYnsVJ2RSMX779tFlY5u3Wz5Q8dxos8f/2ipOPKha9dwHgTj\nQNXHJDZuhHbtbHRT06ap63zve3DccbbOhOM40eWLL6BtW8vFtt12ue9XW2v1166Fbbctn76o4jGJ\nDHzwgc20TmcgwNxN3pNwnOhz5ZU2vykfAwHQpIn1Jvx3vjVVZyQa+vSyuZogPHdT2L7dfImTXtda\nPsLSqwr33GOZX3MlWWuPHtE3Eh6TCIEFCyxFeCZ23tkWJHIcJ7qsWGE9gl69Ctu/e3cb8u7Up+pj\nEj/5iV1UP/tZ+jqLFtXNlUg14c5xnPCZOBEuu8wyOhfClVfa5Ltf/KK0uuKAxyRScPfdNjdi3DhL\n7pWJHj0sZrFwYUWkOY5TAM8+W1yCvh49vCeRiqozEgmf3tix8MwzluX10EMz7yMCBx9c+BNKobgv\nuny41vIRht533oHRo2HYsPz2S9a6664Wo4wyHpOoILNmmRtp7Njc6h9yCEyeXF5NjuMUxoMPwnnn\nWa61QvEV6lJTlTGJdeugY0ebG5FrjGHSJItfvPFGkSIdxyk5Rx0Fl18OJ51UeBtbtkCrVhYAb9Wq\ndNrigMckGvDhhzY3Ip8g9P7721OG53BynGixaRNMm2au42Jo2tTWtp83rzS6GgtVZyRqampYuBB6\n9sxvv223hb59Ye7ccqhKjfuiy4drLR+V1jt9ut3c27TJf9+GWvv1g9mzS6OrHHhMokIUYiTAAlsL\nFpRajeM4xTB1KgwaVJq2Dj7YY48NqcqYxGWXwU472bjofLjySvNV/r//V6BAx3FKztChcMQRcMEF\nxbc1bRp8//u2sl014TGJBhTakzj2WBg/vtRqHMcphpkzixvVlMzAgRazXLmyNO01BqrOSCRiEoVM\n3T/qKHjzTfif/ym5rJS4L7p8uNbyUUm9n35qA0p23rmw/RtqbdbMXE4TJxavrRx4TKJCLFxoo5vy\npWlTW+Lw3ntt+KzjOOEycKCNburQoXRtnnkm3HZb6dqLO1UXk9iyBbbZxi6sTOnBM3H88fDDH8IZ\nZxS2v+M4pSExjL2Ut7FVq8wdvWaNJQysBjwmkcSaNdC6deEGAuCAA6I9TM5xqoFly2zBsFKn927f\n3l7z55emvVWr4Oc/hwceKK0xqxRVZySeeaaG9u2La2P33W1d7HLjvujy4VrLR6X0vvACHH10bmtZ\npyOd1mOOgeeeK7zdWbNg+XJ49FFbSvWmm+Dss229i2II41poVvEjhsynn9rTRzF07Wppwx3HCY+X\nX4YjjyxP20cdBY8/Dj/+cf77bt4Me+9dv2zBAksQ+sADlmMqTpQ9JiEiC4G1QC3wpaoOEpF2wAPA\nLsBC4ExVXRvUHwl8H9gM/FRVnwvK9wPuAbYFnlLVS1IcK2tM4tln4U9/Ku4pYcYMOPdcG3rnOE44\n7L47PPQQ7LNP6dtOjICcNs3cy/kwbRqcfDJ8/jmMGmUPpUOGWM+iXz/473+Lc3eXg7BjErXAYFUd\nqKqJeZFXAS+o6u7Ai8DIQOgewJlAf+BEYLTIVxmWbgXOV9W+QF8ROb4QMQsXWt74YujYET75pLg2\nHMcpnI8/tkR8e+1VnvZ79oRrroG//z3/fV96Cb71LVi71noNQ4ZYeadOtgrmLbdAbW1J5ZaVShgJ\nSXGc04B7g/f3AsFp5FRgnKpuVtWFwHxgkIh0Blqr6rSg3tikffLi2Wdr2HPPQvaso0MHm2yzZUtx\n7WTDfdHlw7WWj0roffllS+hX7OijTFoPPriwASovvZTeDfbgg7bg2WWX5d8uNN55Ego8LyLTRCQx\ncb6Tqi4HUNWPgY5BeTcgeW2opUFZNyB5DMOSoCxvZs+G/fYrZM86mje33sTSpcW14zhOYWS6EZeK\n3XbLPyPs6tWm7YgjUm/fdVdzed91l8Uu4kAljMRhqrofcBJwsYgcjhmOZCoyMGztWli6dDCHHVZ8\nW337lj+l8ODBg8t7gBITJ72utXyUW++aNfDII3DKKcW3lUlrz542A/upp3Jra8kSuPpqS9/TuXP6\nel27muupkAWOwrgWyj66SVWXBX9XiMh4YBCwXEQ6qerywJWU8PAvBZIjBt2DsnTlWzF8+HB6BomZ\n2rZty4ABA746sWPH1tCjBzRvbp8TXbfE9nw+9+kDTz5ZQ7Nmhe3vn/2zfy7s85Qpg/n61+Gjj2r4\n6KPyHW/ixBoOOgjeemswJ52Uvf4hh9SwZAncdVf29gcMgPvvr+HYY8M5nzU1NdwTjMftmS2RnaqW\n7QVsD7QK3rcEJgHHATcCVwblVwI3BO/3AN4CWgC9gPeoG4E1BTMwAjwFnJDieJqJv/xF9RvfmJCx\nTq784Q+ql1ySetuqVSU5hE6YMKE0DVWIOOl1reWj3HqPO071kUdK01Y2rWPGqJ57bvZ2vvxSddtt\nVW+/XXXlyuz1r79e9ec/z01jMuU6t8G9M+V9vNzupk7ARBF5K7jJP6E2pPVG4FgReRc4BrghuMPP\nAR4E5gSGYETwBQAuBsYA84D5qvpMvmJefZWig9YJ0q1gNXeuzdb87LPSHMdxnDqWL4fXXoPjjqvM\n8Xr1sqyw2ViwwFxMP/gBOU3W3Xvv+KQjr6rcTQMGwJgxthRpsbz5Jpx/Prz1Vv3y//s/+NGPzJ/5\nwQfFH8dxnDquucaGsd97b9aqJeG99yxX2/vvZ673+9/DnDm5z6ieMwdOP70ymRtyIdM8iaoyEm3b\nmsUvNi0HWJBq0CD46KO6MlV78rjlFjMgr71moxkcxymeDRugSxebxFpIFudC2LjRJsNt2JB+uG1t\nrd1TJk7Mfd7G+vU2lH7DhrokhWES9mS6SLBmjf0zZ8yoKUl7O+1kk3mSJ8XMnWv/8FNPtdcLL1ga\nkJtvLmzyTCLQFBfipNe1lo9y6a2psdTgpTQQ2bRut53lhpozJ32dDz6w9bXzmdjXsqW98p2UG8a1\nUDVGIrEaXams9jbb2D95zZq6snfftRQBInbBXHQRDBsGl18OI0aU5riOU61MmgSHH1754x59NLz4\n4tblW7bYUNxf/QoOOij/dnfZJbd4R9hUjbvpX/+CO++EJ54o3fH69rUkYP362ec//cncUDffbFkg\nE0m+nn7aeharV5thcRwnfwYPhquughNOqOxxx42Df/yj/r3jvvvsofOII8zV9O67+S98dMYZ8J3v\nwLe/XVK5BeHuJgpf1zoTDXM4LVhguVnAehIzZ8Jpp1ngq0cPmDy5tMd3nLBRhXXryn+cKVPswevQ\nQ8t/rIaceCK88oo95IG5kIcOtfJvf9vikoWsjBeXnkTVGYlS+vQ6dUpvJMB6EuPHm/tp551tJuYu\nu+S+SIr7osuHay0N119vi3glZ0Quh96HHoKf/tR8/6UkF6077GBDbh9+2D5PmmQDUo4/Hi691FzP\nhVCIkfCYRBmpVE8i3Wim666zv4sWWazCcRoDjzxif6+5pnzHWLYM/vMfOPDA8h0jG9/4hrmTX3rJ\nXNcXXWTf/eCDC28zLj2Jss64rvSLpBnX3/ymavLkxAEDVKdNy3H6YY6MHKn6u9/Z+82bVbfZRnXj\nxtR1339fFVQffVS1eXPVtWtLq8VxKs2yZart2ql+/LFqmzY267jUPPSQ/W5Add680refK0uX1ukA\n1SlTim/zjTdU99mn+HZKASHOuA6FLVvg3/+GCy4w/6Hq1q6gUtC8OXz5pb1P+CW33TZ13U6d7O++\n+9pIiKlTS6vFcSrBpk3mWqqtteU4d9jBru2OHUub8HLDBuv933KLpdaGys2NSEXXrpa19Q9/sM8N\nV54rhLj0JBqdkTj33Lop8iefDHvsYVP5mzSxUQil9OklG4lsRqhlSwti9+hhwbdXX83efpR90amI\nk17XWhg/+IFlLBg82Fwvf/2rlSenryiF3uOPtzYnToRzzrFV3lq0KLrZrchHa9Omtg6EKmy/ffHH\nbt/eDM/atbnv4zGJEvDQQ/C3v1mOplGj7MJ64onS9yKgvpF4//3sxxg/3vbZd9/Mk3McJ4osXQpP\nPmk5h155xYZwfuMbtm2XXSzeVgpWr7a0NwCPPmq/4UKDw1FGJB69iUZnJDZtsgBTIpHf+vVw4YWw\n4472OZE2txTk05NIpnfv7LlgwNcRKCeuNX+eeAJOOsnmBb34Ivzzn3Xbeve2PEdQvN7nnrOeiqrl\nNyonYZ/bfI1EGHobnZFI+C0TCwu99pr9LccqcslGIp/RU716efI/J34895zNDQA46qj6T/f77AMz\nZpTmOP/+d10PpbHjPYkQWLgQnn++7iLr2RMee8yyv0L5YhIrV1o+p1zo0MHSeST2TUeUfNG5ECe9\nrjU/Nm+GCRPg619PvX2ffepSXxejd8sWy1BQKSMR9rnt0SP3eVMQjt6yr0wXBg0v5CFDynOcZCOx\napVli8yFpk3N/bVyZeZlDh0nKsycaRlYE6P0GtKtmwVgi11HZeFCS6oX5kimSrLjjvado0yj60lk\no1wxiVWr8ktB3nAiXirC9pfmS5z0utb8mDIFDjkk/fYmTWwi6XvvFaf3vfcsJ1qlCPvctm9v945c\n8ZhEzGjRojgjsXx5eXQ5TqnJZiTAVmtMBK8LZeZMa6daaNeuLidUVKk6I1GOmERtrcUY2rbNfd8e\nPbIPGXz22Zp6ixpFnbD9u/ngWvNjypTsKSj69IH584vTe//9cNZZBe+eN2Gf23x7Ej5PImYkjMSh\nh5qhaJZHhGfXXbMPg739dvP1Ok6Y/Oc/dvPv3z9zvb59i1uO84svbOGuQtZmiCvek4gg5YhJJIbZ\n5kPv3ja3IhNffDEYgLvuyr/9MAjbv5sPrjV3rrgCrr3WBlxkYt99bRhsoXpnzrSHp+22K2j3ggj7\n3HpMopGTHLjOl27dLLtlOlTh9dfh73+HkSPts+NUmo0bLTvAFVdkr7vnnpa/6YsvCjvWf/4DxxxT\n2L5xpXVry1NV6H2kElSdkSjXPIl86dwZPv44/faPP4YNG2o45xzL+5QqjcdDD2Vuo9KE7d/NB9ea\nG7NmWayhefPsdbfbznrIY8fWFHSsadOyB8dLTdjXQZMmFstMXgY5Ex6TiBnFGIkuXdL3JNautZWv\nuna1/C5HH20TmRpy5pnWzn/+U5gGx8nGgw/mN7Ftzz0LH/f/1lswYEBh+8aZdu3yczlVmqozEqWO\nSUyZUti+bdrYK1Wg7/HH4YUXAAYDtvj7pEn16yQmLYmUdt3uYgjbv5sPrjU7qpZg7+yzc9+nRw/Y\nYYfBeR9r7VobEl7p4a9RuA7yiUt4TCJm5NIFT0eih/Dyy1tve/xxy1v/7LP2+cgj4YEHbBRVIvXB\nP/5hyc8yKIvJAAAgAElEQVTeftuGDa5fX7gWx0nF3LnWU95nn9z36d49vzQTCWbMsDUasgXHGyM7\n7FD8TPVyUnVGotQxCYDRoy3vfb707Jn6BzV1qqVhXrCg5qt6t98OkyfX/WCfesrGk++5py3rmFhG\nMkzC9u/mg2vNztSpcMQR9kCTK926wVtv1eR9rLBcTVG4Dlq2hHXrcqvrMYmY0bcvjBgB3/xmXdbZ\nfOjRAxYv3rp81SpLApjMBReYIWrRwnoRr79eN7npvPNg7Nj8j+84mfjgg/zXYeneHVasyP9Y06fD\nwIH579cYaNUqdyMRBqKNaGyliGicvs8zz8BNN1kK5gRffmmrXn3xReonuEcftfUypk+3LmqTJrZE\na9euZlzKsXqXU32sW2fDM2+/3Vajy5XFi+3hJd/U/AMG2LEGDcpvv8bA//yPudpGjAhPg4igqin7\njN6TCJFU/tvVq21IXLoufu/etvRp795mIMAC4D172nBFxykFkyfb3zPOyG+/Ll2sJ5HPqL8tWyz+\nsdde+R2rsRD1nkTVGYko+CATJNxNyZ2f1avrUo6n0ppYsyKx0l6CPfawH1qYROncZsO1ZuaRR2yW\ndcPrLBvNmkG7djV5LaTzySf2YFSKdaPzJQrXQatWuQ888ZhEldGmjf399NO6shUrMmeTTRiJHXao\nXx4FI+EUzrJl0ZpVP3Fi4Qv/dO2aPeVMMkuW2ANTteI9iYgRhXHRCUS2Dl6/+25dPv1UWhMxh4Y3\nlP79wzcSUTq32YiS1tpau7Gef37q7WFoXbwYdt65sH0PPHBwTmu4J3jlFdh998KOVSxRuA7yGd3k\n8ySqkIZxidmzc/PNNpzGHwUj4RTGwoXmbrn7bjj22PD/j59+asuV5rrSYkNyyXCczNixcNFFhR2r\nMeA9iYgRBR9kMj162GilTZvs86xZNvcBMmttuKrdLrukHk5bSaJ2bjMRFa2bN9tNdcAAG7G2116W\nv+j11+vqVFrr1KnQr19+8yOS+fzzGt56K7e6K1bYUNtsa1WUiyhcB/kYCY9JVCHDh1sKjsTIpNmz\n64xEOkaPhhtvrF/Wpo25LaI8c9PZmkQvcvNmS7V9882Wt+v558PTdOmlcNllhe8/cKBdx4nsAJmY\nMMEm7BWTvSDu5BO4DgOfJxEBDj8crrvObhLdull3v0kB5rtPH3jyyfD8u07+1NTAUUfZ0/uBB1rZ\nmDHmp7/nnsrrScy5Wbu2uBQZ559vvaMf/zhzvQsvtEEXl1xS+LHizpQp9v0LzQNXChrNPAkROUFE\n3hGReSJyZdh6SsX221tO+XfesRt8IQYC7Mcdp+VOHUv3/pOf1BkIgF69Cs+kWizjxsF++xWfQ2nI\nEGsrG9W4hkRD8glch0FsjISINAH+ChwP7Al8R0T65dtOFHyQDWnZ0rqbc+fWXyIyX61hG4kontt0\nREHrvHnWa2j4tL3TTvDf/9Z9rpRWVbjhBvjjH4trp6amhhNOsOD1vHnp682fb9d9mJPoonAdeEyi\ndAwC5qvqh6r6JTAOOC1kTSUhnZHIl7CNhJMfL71ka4Lstlv98p12Kiz/UbF89JHFtJJ7NYXSvDmc\nc46trJiOK6+E73638AB5YyHqo5uahS0gD7oByeN3lmCGIy+iMC66IdtvD3/9q72//PK68ny1dusG\nixaVTle+RPHcpiMKWufNM398Q3bc0fJw1daa67FSWl97DQ44oPibdkLv8cdbzyQVn31mqfDDMIbJ\nROE6yCdw7fMkqpTttrOlG6dNq5tRXQjduhWWy98JhwULUmdZbd7cbhy5LmlZKh56CE4rYd/8oIMs\nBfjKlVtvGzPGAvZhpOKIGttuawk9N28OW0lq4tSTWAokzwHtHpTVY/jw4fTs2ROAtm3bMmDAgK+s\nb01NDdOnT+eSYChFwr+XvD2Mz/PnDw7U1/D++3XbR40atZX+TO2tWFHD7NmQWNGu0t8nX71hfk72\n7Yal5803zXef6v/VoQM8+WTNV7OekzWXQ8/ixfDEEzXBKnTFtZcomz69hkMOgdtuG8wvf1m3fZtt\nBnPTTXDJJTXU1IR7PUTlftCqFTzzTA2tWlVGb01NDfcEw+cS98u0qGosXkBT4D1gF6AFMB3o36CO\nZmPChAlZ61SaoUNVLWyounhxXXm+WhctUu3SpbTa8iGK5zYdYWtdtkxVRPXzz1NvP+QQ1YkT7X25\ntW7ZonrMMarXXlua9pL1Tphg3yVBbW3dtb5xY2mOVwxhXwcJunRRXbIke71y6Q3unSnvvbGaJyEi\nJwC3YG6yMap6Q4PtGqfvk2DLFvPNdulivslCu+Bbtljiv48+qkse6ESTCy+0OQnphomeeip8//s2\nlLTcjB0Lt95qczOaldi3kFiX4m9/s7kTixfbEF+IVkLDsOnb19aqD2uOU6Z5EnFyN6GqzwCNbqpY\n06bQqRP85jcWnyimnb32svQORxxROn1OaamttUmPL72Uvk5ihNNtt9lNvKamtAtKzZplaz589hm8\n+KIZpFIbCLDYys9/biOZJk+2CXbf/Cbcd1/pjxVnojzruuoC18l+0yghAr/6Vf2RJYVo7dfPxp+H\nQVTPbSrC1Pr22zbsuU+f9HU6doTlyy39ynvv1XDWWaV78p4/H772NZs0d+SRcO+9pRn2mqDhuf3j\nHy2APXaszSz+8Y+LexgqJVG5ZnMdBhuG3qozEo2dbt3yXzrSqSxz59oTdSa6drXew5df2prm48fb\ncNhUI4XyQRV+9jO46qq6JJEi2fUUS8+eNppr1iwzTE59ojxXouqMRCLSHwcK0ZrNSKxbB9/7nt14\nSu0TbuzntlR88EHqoa/JdOliKSsGD4Zjjx3MAw+YO/GVV4o79iOP2PEvvdRcWnffbb7wUpLu3O6y\nS/bklZUmKtdsrqk5wtAbq5iEk52997YgpGrqSVEXXmhPdJddBttsA9/6VuU1Vjvz52dPjZ1YNvSO\nO+zvmWfCnDk24a3QYPacORY8/ve/6+Ibw4cX1pZTWrwnESGi4oPMhUK0HnaYjSBZtap++aef2roV\nr7xi6ZmHDSt97KKxn9tSMXOmGfNMDB5s+Zu2265O68EHm5EolHHj4LzzLB5RTvw6yJ9cA9dh6PWe\nRCOjSRPr1i9aVH8R+5NPtoWNJkywG0/r1vCLX8CJJ5bfH+3U5513UqfjSEak/v8PLGXGG2+k7yVm\nYssWW4dkwoT89nMqQ5R7ErGaJ5GNuM6TKDWnnAIXXFCXYmHRIth/f/j447oU0H/5i6WohuLmZjj5\nsXGjLVX6+eeF5Ujq0sV6E/muPz1/vi2NGlYKcicz111nywVcd104x28060k4udGnj/mfEzzzDBx3\nXP01Alq2tL+77WYGZcuWymqsVlatsh5CoUn0+vWDd9/Nf79zz7URRk40iXJPouqMRFR8kLlQqNZD\nD4VJk+o+P/+8ZeRMprbW/o4fD/ffDz/4QWEak6mGc1ssq1ZB+/b57ZOstUeP/Ic4P/us9T7Gj89v\nv0Lx6yB/ch3d5PMknJJw2GHw6qtmCFRh4kRbIjWZL76wv/37w5//bEMjkxe6ccrDypVbxxryoXv3\n/DP9PvaYzXhu27bw4zrlxWdcR4iojIvOhUK1dukC7drZpK0PP7Syhq6GAQMsgN2kic2AHTDA0nkU\nQzWc22JZuTL/nkSy1kImS06ZAmeckd8+xeDXQf7k6m4KQ2/VGYlq4bDD4NFH4ac/tTWEG/rADz3U\nAmUJSmEknOx8/DF07lz4/vn2JNats6C1j2CLNh6TiBBR8UHmQjFazzvPckG1agV33pm9/oABll+n\nGKrl3BbDsmXW08uHZK35Gok33oB99y1tcsBs+HWQP567yak4Rx0Fo0bBzTfbylfZGDzYApyVXg2t\n2ijESCTTvbtNlsyVV17JPrvbCZ9cA9dh4PMknK8YMsR810OHhq2k8XL44ZYS/uijC9u/ttbSqWzY\nYMucZmPAABuY4Knjo83ixeYCzucBoFSMHQvDhvk8CScHDj8c3nwzbBWNly+/tJQc++5beBtNmkCH\nDnUZXDOxZg28957dfJxoE1ZMYtMmuOKKzHWqzkhExQeZC5XW2quXZQgtFD+3mbntNrth5zsEtqHW\nTp1srYlsvPmm9STKsZhQJvw6yJ+EuymbI6TUev/wBxg4MHOdqjMSTnqKNRKNldpaS63+2WfFtfPw\nw3WpUIqhUyeLbWRj+vTsNwAnGrRoYb3ExPylSqAKt99ui0JlwmMSzlesWWMzej/9tPC0EY2RGTPq\nhpDOmlXYmgiq1oN45x1bda4YfvlLS6Nyww2Z6w0bZi7ECy4o7nhOZdhxR5g3r7jJlvkwf74NcFm8\nGJo08ZiEkwNt25protjVzxobr70G3/2upS7Za6/C0nUvXWq5s4o1EGA/7ClTstebMaO4+IdTWVq2\nrOys6xdeSD2HqiFVZySi4oPMhTC0FuNyitq5nTPHntxTkatWVXjoIYsl/O1v9vR+4435a3nuOftB\nFkJDrX37Zl8L5IsvLBHgXnsVdsxiiNp1kIkoac0leF0qvevW2XX99a9nr1t1RsLJTGOJS6xfb26h\nYuYIfPGF+YlfeMGe3kVgxAj7/Omn+bU1a5atB1EKune33t7GjenrzJ1rqVi22640x3TKTyVHOF11\nlc2fymWVw6ozElHJ1ZILYWgtxkhE6dw+9pj9XbsWnnpq6+25aL3nHvt7442WCBFssaavfc3Sr+fD\nwoWFp+puqLVJE0vtkSl4PXWqrSESBlG6DrIRJa25GIlS6K2ttZQ9t9xi13M2qs5IOJlpLD2JiRPN\nNTRwIHzjG3Wp0XNl40a49FJ4++2tx5Gfdlr+abffe89WDCwVXbtmNhIvv2yz6J34UKmexGuvWZLJ\nPn1yq191RiJKPshshKG1Z8/4xyT++1/rSZx2mt0smzSxpTuTyaZ11ixbkCmVT/+44+Cll3LX88EH\nNq+h0CR7qbR26ZI5G+z779sCRWEQlesgF6KkNZfAdSn0PvYYnH567vWrzkg4mWkMPYnrrjNfa79+\n9nT2m99YOvRcZikneOut9HMMunaFFSuyT3xKMG2aBb5zSaORK7vvDrNnp99ejHvLCYdWrYqfi5ON\np5+2CXT5GAmfJ+HUY/16G6e9cWM850ps3mzrP0+YYDfSBEceaVlxcx1h9D//Y3GIdJPfdtjBbsTt\n2mVv69e/Nl3XXpvbsXPhiSfgr3+1pIwN2bQJ2rSx/E7JS9Y60WbkSPu/jRxZvmMMG2Y97SefrP/7\n9jWunZxp2dKeePMdvRMVxo61HkSygQC74eezNnSmngTATjtZbyIXZs8ubAJeJvr0MSOVikWLbASU\nG4h40bFjfr3dfFm3zmJp99yT3wNg1RmJKPkgsxGW1i5dbHGcfInCuX3kEbjwwq3Le/Son2Ezk9Yt\nWywmkWkiWj5GYs4c2GOP3OqmIpXWHj3MGKTqOC9cWNogeb5E4TrIlShp7dgxe06uYvROm2YPKzvt\nlN9+VWcknOx07lyYkQib1attVNNJJ229raGRyMTMmVa/TZv0ddq1s+G12di0CRYs2LpnUywtW9or\n1brkH3zg8Yg4Uu6exOTJcMgh+e9XdUYiSuOisxGW1lyeaFIR9rm9/XaLOaS6uffsaTfrBJm0vvKK\n5TzKRJs2ubnk5syBXXfNbeGndKTTuvPO1ptoyKxZsPfehR+vWMK+DvIhSlpzMRLF6J00yY2EUyJ2\n3DF++Zs+/NBGbfzv/6bevvfedvPMZb5EKY1EOfMnJVxODZk5E/bZpzzHdMpHOXsSK1aYkcglDUdD\nqs5IRMkHmY2wtHboUJiRCPPc3nEHfO976ecGtGtnKbYTiypl0jpjRvbZyrkaienTizcS6bSm6kmo\nhm8k/DdWGInfXaYHmUL1/v3vcMoplsQzX6rOSDjZ2XHH1L7uqHLttTY34rzzMtf75jdt6F8mNm+2\nXknv3pnr5Wok3nmn9CObEqQyEh9+aMub5hucdMKneXO7rlatKm27qjBmDJx/fmH7V52RiJIPMhth\naU3Xk1i+3ILD6aiE3g0b6kb0qNpa0f/v/8HVV2d/Yt93X4sRQHqt119vif2yxRByNRILFlhMohgy\nxSQaBuMTyQjDxH9jhZNt1cFC9E6fbr+bQtc5rzoj4WQnVU9ixQobVnnyyeFoAlsjumVLuPlm+7x4\nsU2aA/jtb7Pvv/fe8Prr6WdKT5xoE+7uuit7WzvsYIs0ZeLzz2HJkvKNNErVk1iyJPecPE70KEdc\n4plnzNXUpMC7fdmMhIhcIyJLROTN4HVC0raRIjJfROaKyHFJ5fuJyEwRmScio5LKW4jIuGCfySKy\nc6G6ouSDzEaUYhJPPWX+/sQF/PrrNp8iebnFcuu9+GLLWnnttZa36Oc/t/KmTXObHJRw+8yZk1pr\nYvZyLjf1jh2zz5N47TUzTMWMbIL8YhLr18P22xd3vGLx31jhZDMSheh97jnLN1Yo5e5J3KSq+wWv\nZwBEpD9wJtAfOBEYLfLVT/xW4HxV7Qv0FZHjg/LzgVWq2gcYBfy+zLqrmlQ9ialTLZvqqlX2/sAD\nbS5FOcd1J3PDDfD44/DRR5bj5qCDbM3ogQMzu8CSETEDkG4OyPTp9t332y97W7k88T3yCJx4Ym7a\nCqFzZzPmmzbVla1fb70tJ56Uuiexbp1NojvyyMLbKLeRSPV8dxowTlU3q+pCYD4wSEQ6A61VdVpQ\nbywwJGmfe4P3DwMFrvEVPR9kJqIUk5g2zZ5G1q6tH/xNvqDLpXfxYhveOnmyGYhf/AJOPRW+9S1L\n5JdLTvwEieG9DbVu2GBDBGfNMldSNtL9mBcssHO0fj384x/Zg+m5kO68Nm26dTbYDRvCNxL+Gyuc\nbEYiX70vvWSLXbVqVbimZoXvmhM/EpFzgdeBn6vqWqAbMDmpztKgbDOwJKl8SVBO8HcxgKpuEZE1\nItJeVUs8DsABu0lu3Gg+9W23NbfK3Ll1F9sHH9hN+x//qExPYtIkWxuhVy/7fNFFhbfVvn3q0SPj\nx9sqdp0759bOTjvZd1et7+radVcLkM+YYaNVyp0eI+Fy6t0b5s2De+8tb+/FKS8dO1resFIwfryl\nyD/22OLaKaonISLPBzGExOvt4O8pwGigt6oOAD4G/lSc1PqHLnTHqPkgMxGWVpH6+Zseftie2Fu2\ntJvjv/5lN78996xvJMql9913S7c2Qvv29r0aap0wIb8f0/bbQ7Nm9ReJSSQQnDHDhqEedljxeiHz\neU2OSyQWQgq7J+G/scIpVUxC1dKBP/88/PCHxWkqqiehqrn+rO4AngjeLwV6JG3rHpSlK0/e5yMR\naQq0SdeLGD58OD2DyGPbtm0ZMGDAV120mpoapk+fXu8zENnP06dPD+34XbvCk0/WsNdeMHfuYAYO\ntO1f/zrcdttgevSAZs1qePZZGDq0vHpff30w3/teadpbuxZuvXUwM2cC2PZBgwbz0ENwxx011NTk\n3l7r1jU8/jh897v2+S9/qeGUU+CeewbTrp3Vz6e9dJ8TpNquCosW2eeZM217y5bFHa+ceqP2OWr3\ng6VLYfny4vX++9/QuXMNd90FHTpsvb2mpoZ7gvV5e2YbqaGqZXkBnZPe/wz4Z/B+D+AtoAXQC3iP\nunUtpgCDsJ7CU8AJQfkIYHTw/mwsppHqmOqUhjPOUH3wQdX161V33FH1hResfPJkVVDdtEl1/HjV\nE08sr47PPlNt3Vp19erStPfvf5v+rl1Vn31WdcsW1UmTVPffP/+2Dj5Y9dVX6z6fdZbqvfeWRmeu\n3Hqr6oUX2vuhQ+27PfdcZTU4pWPePNVevYpro7ZWdbfd6n6zuRDcO1Pey8sZk/i9iAwAaoGFwA+D\nu/gcEXkQmAN8CYwIRAJcDNwDbAs8pcGIKGAMcJ+IzAdWBobCKSO9e9u6zG+8YVP5jz7ayg86yNwq\nLVrYyKJS+U/TUVNjsZBC0gmkIpGy+6OP4Pjj4T//qRutlS8NXQMLFkDfvqXRmSs9epj7D+rcXfkE\n8p1o0b27XZu1tYXPa7jqKptTlPjNFkvZRjep6lBV3UdVB6jqEFVdnrTtelXdTVX7q+pzSeVvqOre\nqtpHVX+aVL5JVc8Myg9WGxVVEA27xFEmTK177mmL5RxxhI2eSQRnRepuhD162DyJZcvsczn0zplT\n+NrQqejZ00Ye/eQnNZx+umWNHTnShvfmS0MjsWiRxQhKTabzmohJbNliKUA+/NAC8GHiv7HC2W47\nGziSbph2Nr1PPQUPPGBJKku1sqTPuHZSctBBNnoJ0q+FIGJrN0+cWD4d778Pu+1W2ja3396CeuPG\n2fuBAwsbEZRsJDZtsvkanTqVVms2evUyQ3r66fYUWg4j5VSWXXZJnd03Gxs3WkaEW2+1B7hS4Wtc\nOylRtRnORxxhF166cdZ//KOthPbXv5ZHx557WpqMgw4qT/vFMGqUDQe+5RabiPfd71rvq9Iknhh/\n9jO46abKH98pLd/6Fnz723DWWfntV1MDV1xh7tN8ybTGdbnnSTgxRcTGWGfjyCNtbH45mDzZJocV\nEi+oBB07wpQp9v7eezOviV1O1q0zI37OOeEc3ykthfYkbryxNJM3G1J17qao+SAzEQet++5rAe6N\nG0uv94knYOjQwgN4mSiF1kQ8YPZs61WMHFm8rlRk09qypcUkDjigPMfPlzhctwmiqLVXr7pBCA1J\n1qtq6WpWr7Y04FOnwrBhpddTdUbCKS0tWlgguxxulqlTYdCg0rdbKnr1MjdTIk9/udaNyIVyGFIn\nHL7+dUvKl4233rIHk/btLS71zjvlSe7oMQmnaM4802Zkl9LdsXGjuXOWLMktl1IY1NZa/iSwobSl\nGnLoVDe1tZYO57PPbGRhugWwEj3XIUNscEkxw8QzxST8+cMpmj59YP780rb52GM2ciqqBgLs6T2x\ntoWvKe2UiiZN7AHp448tF1iQyGAramrghBNsUEep5hGl1FO+pqNJFH2Q6YiL1oSRKKXeRx4pbyC2\nVFqPOMJSi3foUJLmUhKX6yBBnPRGVWuXLjb8GyzmlyCh9/PPbS3zSgyWqDoj4ZSeUvckvvjC3Dcn\nnVS6NsvFfvvZrHTHKSW77lq3CNbbb2+9/fzzLXbRpk35tXhMwimaTz6xLK2lWsD9hRds3erJk7PX\ndZzGyKOPwtlnW3qN00+3z2CzqUeOtKHh779fuoy/Pk/CKSs77WRDMFeutEV9iuWf/7Q1eR2nWjn0\nUDMQZ59d97D0+ef2GWwSa6VSwleduymqPshUxEWriLmcHnigpl755s3Z14FuyCef2FPTiBGl05eK\nuJxbiJdWiJfeqGrt3NnSa9x2m+Uae/ZZi0H06FHD5s1167tXgqozEk556NvXlhlN5uGHzW+ajwfw\noYcsDUg5R2s4Thy46CIb3demTd0opgMOqBt2XSk8JuGUhN/9zvyk119vRuGVV+oWX3/9ddh//+xt\n1NbaDO4//cnW03Ycx2ZVT54MzzxjywinmzdRDD5Pwik7hx4Kd95pyyUOGlRnIK65xnoTt9+evY3b\nbrN1oYtdk9dxGhNXXWVrhmzaVB4DkY2qMxJR9UGmIk5ajzkGNm+u4bjjbM2G0aMtAP3rX9uIjMsv\nN99qOqZOtVEb//xn6fLgZyJO5zZOWiFeeuOkFcLR66ObnJJx2WVwyCFbp6c47jjrXTz/vPlWt922\n/nbVulTg/fpVRqvjOLnhMQmnIvz+9zB+vPlW5861LJdHHw3Ll1v6gcMPt3r+73OcyuPzJJzQufhi\nmDDB3h97rCXuGzzY8s+AxTQ8WO040cNjEhEmTlohs96WLeHppy33/U9/Cj/5iRmIE0+EMWNsNNQ1\n11RMaqzObZy0Qrz0xkkreEzCqQLatrXYBcDVV9sMbV8LwXGii8ckHMdxqhyfJ+E4juMURNUZiTj5\nIOOkFeKl17WWjzjpjZNWCEdv1RkJx3EcJ3c8JuE4jlPleEzCcRzHKYiqMxJx8kHGSSvES69rLR9x\n0hsnreAxCcdxHCdieEzCcRynyvGYhOM4jlMQVWck4uSDjJNWiJde11o+4qQ3TlrBYxKO4zhOxPCY\nhOM4TpXjMQnHcRynIIoyEiLyLRGZJSJbRGS/BttGish8EZkrIsclle8nIjNFZJ6IjEoqbyEi44J9\nJovIzknbhgX13xWRocVojpMPMk5aIV56XWv5iJPeOGmFeMYk3gZOB15KLhSR/sCZQH/gRGC0yFfL\n298KnK+qfYG+InJ8UH4+sEpV+wCjgN8HbbUDfgUcCBwEXCMiOxQqePr06YXuWnHipBXipde1lo84\n6Y2TVghHb1FGQlXfVdX5QENf1mnAOFXdrKoLgfnAIBHpDLRW1WlBvbHAkKR97g3ePwwcHbw/HnhO\nVdeq6hrgOeCEQjWvWbOm0F0rTpy0Qrz0utbyESe9cdIK4egtV0yiG7A46fPSoKwbsCSpfElQVm8f\nVd0CrBWR9hnachzHccpM1uVLReR5oFNyEaDAL1X1iXIJY+veSUlYuHBhOZotC3HSCvHS61rLR5z0\nxkkrhKRXVYt+AROA/ZI+XwVcmfT5GSye0BmYm1R+NnBrcp3gfVPgk6Q6tyXtcxtwVhod6i9/+ctf\n/sr/le7+nrUnkQfJT/6PA/8QkZsx19BuwFRVVRFZKyKDgGnAUODPSfsMA14Dvg28GJQ/C1wXBKub\nAMdiRmgr0o3zdRzHcQqjKCMhIkOAvwAdgCdFZLqqnqiqc0TkQWAO8CUwImmW28XAPcC2wFOq+kxQ\nPga4T0TmAyuxHgSqulpEfge8jlm83wQBbMdxHKfMNKoZ147jOE5p8RnXjuM4TlrcSDiO4zhpcSPh\nOI7jpMWNhOM4VY+I/CpsDblSaa1VayTidFFAvPS61vIRJ71x0gpcELaAPKio1qod3SQii1R15+w1\no0Gc9LrW8hEnvVHTKiKfptsEbKeqpZw3VhRR0hqZk1IOsp3oSmrJhTjpda3lI05646QVWAMcqKrL\nG6qvOvEAAAXMSURBVG4QkcUp6odJZLQ2aiNBhE50jsRJr2stH3HSGyetY4FdgK20Av+ssJZsREZr\nY49JJE50KqJ2UUC89LrW8hEnvbHRqqpXq+rUNNuurLSeTERJa9XGJBzHqT6Cxc8GUbfcwFKCvHLh\nqcoPEemnqu9U7HgxOjclQURGqOrosHXkgoi0AvoCC6KWr0pEWgBfJn5cInIUsB8wR1WfDlVcA0Rk\nH1WdGbaOfAiW7/1UVdeISE/gAOAdVZ0VqrA0iMgBQA9gCzCvkjexXAmWUR6NLYK2NCjujiUgHaGq\nz4WlLR8qPSCgURsJEbm0YREwEvhfAFW9qeKiMiAio1V1RPD+a1h3/X3sIv6hqj4Vpr5kRGQGMDhI\nwHg5toztU8CRwBuqmjJTbxiIyBZgATAOuF9V54QsKSMichXwQ2AT8EfgMmAScDAwJkrXrYgcCfwJ\ni03sj+lshyX2PFdVIxOXEJG5wInBapnJ5b2wZKP9QxGWAhH5c7pNwDBVbVMpLY09cP0b7MY1m7pU\n5k2B1qEpyszBSe9/BwxR1TdFpDfwIPZdokJTVV0dvD8LOFxVN4rIDcCbpEnnHhIzgXOB7wCPi8h6\n4H5sid2FYQpLw7nAHsD2wEKgt6quEJGWWCr9yBgJbD364wJ9vYCbVPUwETkWy+x8XLjy6tGM+itj\nJlgKNK+wlmycB/wce1BoyHcqKaSxG4k9saeclliK8Q0iMkxVfxOyrlzYQVXfBFDVBSIStUEGn4rI\nXoH7479Y6veN2DUVNa0a6Pwl8MtgPZOzgYlB1/3QcOVtxZbA4H6BndOVAKq63lzqkaKpqq4I3i8i\nCGKr6vMiMio8WSm5C5gmIuOoWxK5B3YtjAlNVWqmAbNU9dWGG0Tk15UU0qjdTQlE5DTgCuBm4Peq\n2jtkSSkRkQ3Ae1ivpyewc+DOaQLMVNW9wtSXjIjsA9wHzAiKDgNeBvbGniYjM7JFRN5S1YEpygU4\nQlVfCkFWWkTkHqAF9nCzAdiMrdx4NNBaVc8MT119ROQubJ2XF4FTgaWqeqmIbA+8qar9QhXYABHZ\nA9OZHLh+PGouSBFpD3yuqhtC11INRgK+CgJfgy2RekTYelIhIg2HEi5T1S9EpAN2M3s0DF3pEJGm\nmDuhL3Vd+WcjGGQ/J0pGKxsi0gxbnVGBh7Glf7+DPan/n6quD1FePUSkOfADzD02A7hLVbeIyHZA\nR1X9MFSBTtFUjZFwHKe6CZZAHgkMATpiRvgT4F/ADVF6uImS1qj5jkuKiLQSkd+KyOxgbe0VIjJF\nRIaHrS0VcdLrWstHBr3DwtbWkCSts2Jwbh8EVmOj8tqr6o7AUUHZg6Eq25rIaG3UPQkR+RfwGPAC\ncCbm4x0HXI35Tn8RorytiJNe11o+4qQ3ZlrfVdXd890WBpHSqqqN9gXMaPB5WvC3CTYxKXSNcdXr\nWl1vDLU+hw1g6ZRU1gm4EnghbH1R1dqo3U3A+mBSGiJyKrAKQFVrqZs3ESXipNe1lo846Y2T1rOA\nHYGXRGS1iKwCaoD2WC8oSkRHa9gWs8zWeB9gKubHmwj0Dcp3An4Str4463WtrjduWgNd/YCvA60a\nlJ8Qtraoam3Uk+nU8vUMSlG+QkQ+C0FSRuKk17WWjzjpjZNWEfkJcDEwF7hTRH6qqv8KNv8vNhcl\nEkRJa6MOXGdCIrZqVjbipNe1lo846Y2aVhF5GzhEVdeJJU18GLhPVW9JN+EyLKKktVH3JEQkXeZP\nwYJAkSJOel1r+YiT3jhpBZqo6joAVV0oIoOBh4NJrFGLn0RGa6M2EthFejzmL01GgK1yokSAOOl1\nreUjTnrjpHW5iAxQ1ekAwVP6yVhOp73DlbYVkdHa2I3Ek1jQZ3rDDSJSU3k5WYmTXtdaPuKkN05a\nh2J5sL5CVTcDQ0Xkb+FISktktFZtTMJxHMfJTmOfJ+E4juMUgRsJx3EcJy1uJBzHcZy0uJFwHMdx\n0uJGwnEcx0nL/wfTMzNCHrj1NQAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0xa9a6a14c>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plot( nfp_dev )"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAYkAAAEaCAYAAADkL6tQAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nAAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0xa97c232c>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "#  Zoom in on deviations post-2000:\n",
    "plot( nfp_dev['2000':])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "From 2008 through 2011, over 11 million workers were removed from NFP. \n",
    "As of April 2016 the recovery has been gradual, but still sub-trend. "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Forecasting NFP as univariate time-series\n",
    "\n",
    "*Presumably* examining the first difference series gives insight into \n",
    "how the economy is developing over the short-term. To forecast NFP, \n",
    "it is instructive to go back to the original series and use \n",
    "the Holt-Winters method."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "         Forecast\n",
       "0   143915.000000\n",
       "1   144205.863284\n",
       "2   144428.671615\n",
       "3   144651.479947\n",
       "4   144874.288279\n",
       "5   145097.096610\n",
       "6   145319.904942\n",
       "7   145542.713274\n",
       "8   145765.521605\n",
       "9   145988.329937\n",
       "10  146211.138268\n",
       "11  146433.946600\n",
       "12  146656.754932"
      ]
     },
     "execution_count": 15,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#  Forecast 12 months into the future:\n",
    "nfp_fc_holt = forecast( nfp, 12 )\n",
    "nfp_fc_holt"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Forecast    290.863284\n",
       "dtype: float64"
      ]
     },
     "execution_count": 16,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#  Forecast NFP change for next month's release:\n",
    "nfp_fc_holt.iloc[1] - nfp_fc_holt.iloc[0]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "A large difference between such forecasts\n",
    "and the \"street\" expectations can cause\n",
    "major shifts in the financial markets\n",
    "on release day, especially the bond market.\n",
    "\n",
    "This is especially true when the Federal Reserve\n",
    "places more emphasis on labor market conditions\n",
    "than economic activity in making their policy decision.\n",
    "Note that employment is one of two mandates\n",
    "which drives monetary policy."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "                Y\n",
      "count  196.000000\n",
      "mean   -61.894694\n",
      "std    229.548441\n",
      "min   -951.910000\n",
      "25%   -154.160000\n",
      "50%     -0.910000\n",
      "75%     89.840000\n",
      "max    393.090000\n"
     ]
    }
   ],
   "source": [
    "#  Show recent dispersion around baseline expection of 129:\n",
    "stat( nfp_dif_base['2000':] )"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Our point forecast is unfortunately clouded by a rather\n",
    "large standard deviation of approximately 230.\n",
    "\n",
    "\n",
    "### Forecast NFP dif by local range\n",
    "\n",
    "An *alternative* is to notice how nfp_dif has been range bound since 2011. \n",
    "The mean can serve as a point forecast for the change in NFP, \n",
    "and the standard deviation can estimate the standard error."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "                Y\n",
      "count   64.000000\n",
      "mean   204.296875\n",
      "std     74.896263\n",
      "min     42.000000\n",
      "25%    149.000000\n",
      "50%    207.500000\n",
      "75%    266.000000\n",
      "max    346.000000\n"
     ]
    }
   ],
   "source": [
    "stat( nfp_dif['2011':] )"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "2015-11-12:  Our forecasted NFP change for Nov 2015 = **+205,000** +/- 77,000 at 1 std.\n",
    "(So Fed would be discouraged by a change < +128,000 which happens to be \n",
    "our computed baseline trend.) \n",
    "The street mean prediction for Oct 2015 \n",
    "was +185,000, but the actual turned out to be +271,000 -- whereupon \n",
    "Bill Gross declared 100% chance of December rate hike,\n",
    "which in fact did transpire: indicating the divergence\n",
    "of Fed policy from both the ECB and BOJ.\n",
    "\n",
    "2016-04-27:  Our forecasted NFP change for April 2016 = **+205,000** +/- 75,000 at 1 std.\n",
    "There is street chatter of a recession,\n",
    "so perhaps a negative 2 std change would be convincing,\n",
    "that is to say, less than +55,000 NFP announcement.\n",
    "\n",
    "Interestingly the street expectation for April 2016 was also +205,000,\n",
    "and the actual released figure was +160,000 (thus -0.60 std from street expectation).\n",
    "\n",
    "2016-05-20: Our forecasted NFP change for May 2016 = **+204,000** +/- 75,000 at 1 std."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Measurement sampling error\n",
    "\n",
    "So far we have looked at the raw data, but let us briefly look at\n",
    "the methodology which produces the NFP numbers by the\n",
    "Bureau of Labor Statistics (BLS).\n",
    "\n",
    "BLS uses the Current Employment Statistics (CES) program,\n",
    "also known as the payroll survey or the establishment survey,\n",
    "which is a monthly survey of approximately 146,000 businesses\n",
    "and government agencies representing approximately\n",
    "623,000 worksites throughout the United States.\n",
    "See http://www.bls.gov/bls/empsitquickguide.htm\n",
    "\n",
    "The survey is merely a statistical sampling of the entire working population\n",
    "and therefore subject to measurement error.\n",
    "Currently for NFP, the Bureau estimates the root mean squared error\n",
    "to be approximately 62,800 with respect to the monthly level.\n",
    "See http://www.bls.gov/web/empsit/cestn.htm#section1c\n",
    "\n",
    "Thus measurement sampling error at the source \n",
    "is consistent with our estimated forecasting errors,\n",
    "and should be regarded as the practical minimum error."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Interim remarks on signal and noise\n",
    "\n",
    "Examination of the raw data and the sampling techniques\n",
    "reveal that much of what moves the market upon\n",
    "the release of monthly NFP is largely based on\n",
    "noise. That is to say, reading into any noisy changes\n",
    "within the error bounds of a single monthly release\n",
    "is statistically unsound.\n",
    "\n",
    "NFP as an indicator of economic activity obviously involves *lag*.\n",
    "A rear view mirror is only somewhat useful\n",
    "in going forward fast on the freeway.\n",
    "\n",
    "A *leading* indicator which correlates with NFP is\n",
    "the stock market, as demonstrated in the next section.\n",
    "(We tried corporate earnings as a proxy of economic activity,\n",
    "but the fit with SPX is comparable, moreover,\n",
    "SPX can be readily accessed in real-time.)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## NFP relationship to economic activity\n",
    "\n",
    "Here we explore the relationship between a labor market statistic NFP and \n",
    "US economic activity using the SPX (S&P 500 index) as proxy.\n",
    "\n",
    "When the stock market is booming, it is reasonable to expect more corporate workers \n",
    "on the payroll. There is high correlation, but the data shows that \n",
    "something more than a linear regression model is needed\n",
    "to capture the volatile dynamics for short-term forecasting."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      " ::  S&P 500 prepend successfully goes back to 1957.\n"
     ]
    }
   ],
   "source": [
    "#  Resampled to monthly, the NFP frequency:\n",
    "spx = get( m4spx )"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "#  Data transformed logarithmically:\n",
    "lnfp = np.log( nfp )\n",
    "lspx = np.log( spx )"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "#  Define start and end:\n",
    "start = '1980'\n",
    "end   = '2020'"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      " ::  FIRST variable:\n",
      "count    436.000000\n",
      "mean      11.678566\n",
      "std        0.145405\n",
      "min       11.393793\n",
      "25%       11.583956\n",
      "50%       11.736881\n",
      "75%       11.795143\n",
      "max       11.876978\n",
      "Name: Y, dtype: float64\n",
      "\n",
      " ::  SECOND variable:\n",
      "count    437.000000\n",
      "mean       6.422434\n",
      "std        0.876755\n",
      "min        4.632396\n",
      "25%        5.722735\n",
      "50%        6.784095\n",
      "75%        7.157229\n",
      "max        7.657330\n",
      "Name: Y, dtype: float64\n",
      "\n",
      " ::  CORRELATION\n",
      "0.987661111947\n",
      "                            OLS Regression Results                            \n",
      "==============================================================================\n",
      "Dep. Variable:                      Y   R-squared:                       0.975\n",
      "Model:                            OLS   Adj. R-squared:                  0.975\n",
      "Method:                 Least Squares   F-statistic:                 1.726e+04\n",
      "Date:                Wed, 25 May 2016   Prob (F-statistic):               0.00\n",
      "Time:                        16:04:54   Log-Likelihood:                 1030.9\n",
      "No. Observations:                 436   AIC:                            -2058.\n",
      "Df Residuals:                     434   BIC:                            -2050.\n",
      "Df Model:                           1                                         \n",
      "Covariance Type:            nonrobust                                         \n",
      "==============================================================================\n",
      "                 coef    std err          t      P>|t|      [95.0% Conf. Int.]\n",
      "------------------------------------------------------------------------------\n",
      "Intercept     10.6259      0.008   1314.149      0.000        10.610    10.642\n",
      "X              0.1640      0.001    131.384      0.000         0.162     0.166\n",
      "==============================================================================\n",
      "Omnibus:                       10.493   Durbin-Watson:                   0.076\n",
      "Prob(Omnibus):                  0.005   Jarque-Bera (JB):               12.151\n",
      "Skew:                           0.275   Prob(JB):                      0.00230\n",
      "Kurtosis:                       3.605   Cond. No.                         49.1\n",
      "==============================================================================\n",
      "\n",
      "Warnings:\n",
      "[1] Standard Errors assume that the covariance matrix of the errors is correctly specified.\n"
     ]
    }
   ],
   "source": [
    "stat2(lnfp[start:end].Y, lspx[start:end].Y, intercept=True)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "collapsed": true
   },
   "source": [
    "Given some lag $n > 0$, the regression using log transformed data implies\n",
    "(these equations may not render properly when viewed via GitHub):\n",
    "\n",
    "$$ \\eqalign{ \\log N_t &= b \\log S_t + c  \\cr        \n",
    "   \\log {N_t \\over N_{t-n}} &= b \\log {S_t \\over S_{t-n}} \\cr\n",
    "   \\Delta^\\%_n(N_t) &\\approx b \\Delta^\\%_n(S_t)  \\qquad\\hbox{for some  } n.\\cr }$$\n",
    "   \n",
    "Noting for the third equation that log(1+ψ)≈ψ, for small ψ\n",
    "where ψ can be expressed in percentage terms.\n",
    "The delta here signifies percentage difference over n lags."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Approximate interpretation (where start='1980')\n",
    "\n",
    "The result appears almost perfect:\n",
    "correlation between the two variables is 0.99,\n",
    "and R-squared for the regression is 0.98\n",
    "(the *start* value was fortunate).\n",
    "\n",
    "Since the coefficient b is about 0.164, that implies the NFP will roughly change\n",
    "by one sixth of the change observed in the stock market.\n",
    "This is helpful because we can observe SPX in real-time,\n",
    "and thus it would seem we can forecast NFP accurately in lagged time.\n",
    "\n",
    "We note that our monthly resampling of daily SPX takes the\n",
    "monthly median for robustness (not the mean)."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "                  Y\n",
       "T                  \n",
       "2015-11-01  2086.59\n",
       "2015-12-01  2056.50\n",
       "2016-01-01  1906.90\n",
       "2016-02-01  1915.45\n",
       "2016-03-01  2027.22\n",
       "2016-04-01  2080.73\n",
       "2016-05-01  2057.14"
      ]
     },
     "execution_count": 31,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "tail( spx )"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "                 Y\n",
       "T                 \n",
       "2015-10-01  142595\n",
       "2015-11-01  142875\n",
       "2015-12-01  143146\n",
       "2016-01-01  143314\n",
       "2016-02-01  143547\n",
       "2016-03-01  143755\n",
       "2016-04-01  143915"
      ]
     },
     "execution_count": 32,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "tail( nfp )"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "For May 2016, the SPX monthly percentage is -1.10% thus we can\n",
    "forecast -1.10\\*0.164 = -0.18% for NFP which translates to\n",
    "a loss of -259,000 NFP jobs.\n",
    "\n",
    "We duly note that our baseline expectation of +129,000 NFP jobs\n",
    "at the current level equates to +0.55% monthly SPX increase (+6.56% annualized).\n",
    "\n",
    "Next we examine respective annual volatility\n",
    "(third in the list from georet() output)."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[8.08, 8.91, 12.87, 12, 437, '1980-01-01', '2016-05-01']"
      ]
     },
     "execution_count": 33,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#  SPX geometric mean return\n",
    "georet( spx[start:end], 12 )"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[1.27, 1.27, 0.65, 12, 436, '1980-01-01', '2016-04-01']"
      ]
     },
     "execution_count": 34,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#  NFP geometric growth rate:\n",
    "georet( nfp[start:end], 12 )"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "One can surmise the problem of forecasting NFP using the stock market\n",
    "over the very short-term: SPX annualized volatility of 12.87%\n",
    "far exceeds NFP annualized volatility of 0.65%.\n",
    "It is 19.8 times larger!\n",
    "The former partially reflects speculative activity,\n",
    "while the latter will be more rigid\n",
    "since employment is not instanteous transactionally.\n",
    "\n",
    "It is interesting to note how the geometric growth rate of NFP\n",
    "is 0.157 of the SPX geometric mean return, which is consistent\n",
    "with our regression result.\n",
    "\n",
    "It is a long-term result and should not be solely relied upon\n",
    "for short-term forecasting where trading is involved\n",
    "due to the relative magnitudes of the volatilities.\n",
    "\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Closing remarks\n",
    "\n",
    "We have focused on forecasting the monthly change in NFP\n",
    "using a variety of lens. Here are some\n",
    "point forecasts for May 2016:\n",
    "\n",
    "- Baseline expectation (starting 1939): +129,000\n",
    "- Holt-Winters method: +291,000\n",
    "- Visual selection of local range: +204,000\n",
    "- Regression against S&P 500 (SPX since 1980): -259,000\n",
    "\n",
    "The ideal point forecast would a weighted combination of\n",
    "different techniques. Taking the simple average for the four above:\n",
    "\n",
    "- Mean point forecast: +91,250\n",
    "\n",
    "Clearly the regression with the stock market\n",
    "in this instance greatly reduces the expected NFP number to be released.\n",
    "\n",
    "But this exercise does not end here, for we must consider\n",
    "the difference from the street expectation\n",
    "to assess possible impact on the markets\n",
    "and the Fed's monetary policy.\n",
    "\n",
    "The standard error is relatively large.\n",
    "From the data alone, we observe the standard deviation of 74,900\n",
    "of which the survey sampling error acknowledged by the BLS is 62,800.\n",
    "\n",
    "Examining the NFP data, much more volatility is involved than\n",
    "what is commonly understood in the financial news.\n",
    "We shall later examine employment data in the context of a\n",
    "large-scale dynamic economic model with real-time components,\n",
    "which should improve our forecasting results.\n"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 2",
   "language": "python",
   "name": "python2"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 2
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython2",
   "version": "2.7.11"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 0
}