{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# 11.5 Periods and Period Arithmetic（周期和周期运算）\n",
    "\n",
    "Periods（周期）表示时间跨度（timespans），比如天，月，季，年。Period类表示的就是这种数据类型，构建的时候需要用字符串或整数，以及一个频度（关于频度的代码可以看11.4中的表格）："
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "import numpy as np\n",
    "import pandas as pd"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Period('2007', 'A-DEC')"
      ]
     },
     "execution_count": 2,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "p = pd.Period(2007, freq='A-DEC')\n",
    "p"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "在这个李自力，Period对象代表了整个2007年一年的时间跨度，从1月1日到12月31日。在Period对象上进行加减，会有和对频度进行位移（shifting）一样的效果："
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Period('2012', 'A-DEC')"
      ]
     },
     "execution_count": 3,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "p + 5"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Period('2005', 'A-DEC')"
      ]
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "p - 2"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "如果两个周期有相同的频度，二者的区别就是它们之间有多少个单元（units）："
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "7"
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "pd.Period('2014', freq='A-DEC') - p"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "固定范围的周期（Regular ranges of periods）可以通过period_range函数创建："
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "PeriodIndex(['2000-01', '2000-02', '2000-03', '2000-04', '2000-05', '2000-06'], dtype='int64', freq='M')"
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "rng = pd.period_range('2000-01-01', '2000-06-03', freq='M')\n",
    "rng"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "PeriodIndex类能存储周期组成的序列，而且可以作为任何pandas数据结构中的轴索引（axis index）："
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "2000-01    0.439035\n",
       "2000-02   -0.231125\n",
       "2000-03   -1.085106\n",
       "2000-04   -1.909902\n",
       "2000-05    1.478810\n",
       "2000-06    0.656713\n",
       "Freq: M, dtype: float64"
      ]
     },
     "execution_count": 8,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "pd.Series(np.random.randn(6), index=rng)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "如果我们有字符串组成的数组，可以使用PeriodIndex类："
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "values = ['2001Q3', '2002Q2', '2003Q1']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "PeriodIndex(['2001Q3', '2002Q2', '2003Q1'], dtype='int64', freq='Q-DEC')"
      ]
     },
     "execution_count": 10,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "index = pd.PeriodIndex(values, freq='Q-DEC')\n",
    "index"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# 1 Period Frequency Conversion（周期频度转换）\n",
    "\n",
    "通过使用asfreq方法，Periods和PeriodIndex对象能被转换为其他频度。例如，假设我们有一个年度期间（annual period），并且想要转换为月度期间（monthly period），做法非常直观："
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Period('2007', 'A-DEC')"
      ]
     },
     "execution_count": 11,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "p = pd.Period('2007', freq='A-DEC')\n",
    "p"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Period('2007-01', 'M')"
      ]
     },
     "execution_count": 12,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "p.asfreq('M', how='start')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Period('2007-12', 'M')"
      ]
     },
     "execution_count": 13,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "p.asfreq('M', how='end')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "我们可以认为`Period('2007', freq='A-DEC')`是某种指向时间跨度的光标，而这个时间跨度被细分为月度期间。可以看下面的图示：\n",
    "\n",
    "![](http://oydgk2hgw.bkt.clouddn.com/pydata-book/bomb8.png)\n",
    "\n",
    "如果一个财政年度（fiscal year）是在1月结束，而不是12月，那么对应的月度期间会不一样："
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Period('2007', 'A-JUN')"
      ]
     },
     "execution_count": 14,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "p = pd.Period('2007', freq='A-JUN')\n",
    "p"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Period('2006-07', 'M')"
      ]
     },
     "execution_count": 15,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "p.asfreq('M', 'start')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Period('2007-06', 'M')"
      ]
     },
     "execution_count": 16,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "p.asfreq('M', 'end')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "当我们转换高频度为低频度时，pandas会根据 subperiod（次周期；子周期）的归属来决定superperiod（超周期；母周期）。例如，在A-JUN频度中，月份Aug-2007其实是个2008周期的一部分："
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "p = pd.Period('Aug-2007', 'M')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Period('2008', 'A-JUN')"
      ]
     },
     "execution_count": 18,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "p.asfreq('A-JUN')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "整个PeriodIndex对象或时间序列可以被转换为一样的语义（semantics）："
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "rng = pd.period_range('2006', '2009', freq='A-DEC')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "2006    0.391629\n",
       "2007    0.497413\n",
       "2008   -1.685639\n",
       "2009    0.939885\n",
       "Freq: A-DEC, dtype: float64"
      ]
     },
     "execution_count": 20,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "ts = pd.Series(np.random.randn(len(rng)), index=rng)\n",
    "ts"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "2006-01    0.391629\n",
       "2007-01    0.497413\n",
       "2008-01   -1.685639\n",
       "2009-01    0.939885\n",
       "Freq: M, dtype: float64"
      ]
     },
     "execution_count": 21,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "ts.asfreq('M', how='start')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "这里，年度周期可以用月度周期替换，对应的第一个月也会包含在每个年度周期里。如果我们想要每年的最后一个工作日的话，可以使用'B'频度，并指明我们想要周期的结尾："
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "2006-12-29    0.391629\n",
       "2007-12-31    0.497413\n",
       "2008-12-31   -1.685639\n",
       "2009-12-31    0.939885\n",
       "Freq: B, dtype: float64"
      ]
     },
     "execution_count": 22,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "ts.asfreq('B', how='end')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# 2 Quarterly Period Frequencies（季度周期频度）\n",
    "\n",
    "季度数据经常出现在会计，经济等领域。大部分季度数据都与财政年度结束日（fiscal year end）相关，比如12月最后一个工作日。因此，根据财政年度结束的不同，周期2012Q4也有不同的意义。pandas支持所有12个周期频度，从Q-JAN到Q-DEC："
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Period('2012Q4', 'Q-JAN')"
      ]
     },
     "execution_count": 23,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "p = pd.Period('2012Q4', freq='Q-JAN')\n",
    "p"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "如果是财政年度结束日在一月份，那么2012Q4代表从11月到1月，可以用日频度查看。可以看下面的图示帮助理解：\n",
    "\n",
    "![](http://oydgk2hgw.bkt.clouddn.com/pydata-book/u4btd.png)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Period('2011-11-01', 'D')"
      ]
     },
     "execution_count": 24,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "p.asfreq('D', 'start')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Period('2012-01-31', 'D')"
      ]
     },
     "execution_count": 25,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "p.asfreq('D', 'end')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "因此，做些简单的周期运算也是可能的，例如，获得每个季度的，第二个到最后一个工作日的，下午4点的时间戳："
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Period('2012-01-30 16:00', 'T')"
      ]
     },
     "execution_count": 26,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "p4pm = (p.asfreq('B', 'e') - 1).asfreq('T', 's') + 16 * 60\n",
    "p4pm"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Timestamp('2012-01-30 16:00:00')"
      ]
     },
     "execution_count": 27,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "p4pm.to_timestamp()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "还可以用period_range产生季度范围数据。运算方法也一样："
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "rng = pd.period_range('2011Q3', '2012Q4', freq='Q-JAN')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "2011Q3    0\n",
       "2011Q4    1\n",
       "2012Q1    2\n",
       "2012Q2    3\n",
       "2012Q3    4\n",
       "2012Q4    5\n",
       "Freq: Q-JAN, dtype: int64"
      ]
     },
     "execution_count": 29,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "ts = pd.Series(np.arange(len(rng)), index=rng)\n",
    "ts"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "new_rng = (rng.asfreq('B', 'e') - 1).asfreq('T', 's') + 16 * 60"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "2010-10-28 16:00:00    0\n",
       "2011-01-28 16:00:00    1\n",
       "2011-04-28 16:00:00    2\n",
       "2011-07-28 16:00:00    3\n",
       "2011-10-28 16:00:00    4\n",
       "2012-01-30 16:00:00    5\n",
       "dtype: int64"
      ]
     },
     "execution_count": 32,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "ts.index = new_rng.to_timestamp()\n",
    "ts"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# 3 Converting Timestamps to Periods (and Back)（时间戳与周期相互转换）\n",
    "\n",
    "用时间戳作为索引的Series和DataFrame对象，可以用to_period方法转变为周期："
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "rng = pd.date_range('2000-01-01', periods=3, freq='M')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "2000-01-31    1.556049\n",
       "2000-02-29   -0.708661\n",
       "2000-03-31   -0.154767\n",
       "Freq: M, dtype: float64"
      ]
     },
     "execution_count": 35,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "ts = pd.Series(np.random.randn(3), index=rng)\n",
    "ts"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "2000-01    1.556049\n",
       "2000-02   -0.708661\n",
       "2000-03   -0.154767\n",
       "Freq: M, dtype: float64"
      ]
     },
     "execution_count": 37,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "pts = ts.to_period()\n",
    "pts"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "因为周期是不重复的时间跨度（non-overlapping timespans），一个时间戳只能属于一个有指定频度的单独周期。尽管默认情况下新的PeriodIndex的频度会从时间戳中来推测，但我们也可以自己设定想要的频度。结果中有重复的周期也没有关系："
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 39,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "DatetimeIndex(['2000-01-29', '2000-01-30', '2000-01-31', '2000-02-01',\n",
       "               '2000-02-02', '2000-02-03'],\n",
       "              dtype='datetime64[ns]', freq='D')"
      ]
     },
     "execution_count": 39,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "rng = pd.date_range('1/29/2000', periods=6, freq='D')\n",
    "rng"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 41,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "2000-01-29    1.115254\n",
       "2000-01-30   -1.813124\n",
       "2000-01-31    0.970670\n",
       "2000-02-01    1.306337\n",
       "2000-02-02    0.673274\n",
       "2000-02-03   -0.105436\n",
       "Freq: D, dtype: float64"
      ]
     },
     "execution_count": 41,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "ts2 = pd.Series(np.random.randn(6), index=rng)\n",
    "ts2"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 42,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "2000-01    1.115254\n",
       "2000-01   -1.813124\n",
       "2000-01    0.970670\n",
       "2000-02    1.306337\n",
       "2000-02    0.673274\n",
       "2000-02   -0.105436\n",
       "Freq: M, dtype: float64"
      ]
     },
     "execution_count": 42,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "ts2.to_period('M')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "想转换回时间戳的话，使用to_timestamp:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 43,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "2000-01-29    1.115254\n",
       "2000-01-30   -1.813124\n",
       "2000-01-31    0.970670\n",
       "2000-02-01    1.306337\n",
       "2000-02-02    0.673274\n",
       "2000-02-03   -0.105436\n",
       "Freq: D, dtype: float64"
      ]
     },
     "execution_count": 43,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "pts = ts2.to_period()\n",
    "pts"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 44,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "2000-01-29    1.115254\n",
       "2000-01-30   -1.813124\n",
       "2000-01-31    0.970670\n",
       "2000-02-01    1.306337\n",
       "2000-02-02    0.673274\n",
       "2000-02-03   -0.105436\n",
       "Freq: D, dtype: float64"
      ]
     },
     "execution_count": 44,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "pts.to_timestamp(how='end')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# 4 Creating a PeriodIndex from Arrays（从数组中创建一个周期索引）\n",
    "\n",
    "有固定频度的数据集，有时会在很多列上存储时间跨度信息。例如，在下面的宏观经济数据及上，年度和季度在不同的列："
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 46,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>year</th>\n",
       "      <th>quarter</th>\n",
       "      <th>realgdp</th>\n",
       "      <th>realcons</th>\n",
       "      <th>realinv</th>\n",
       "      <th>realgovt</th>\n",
       "      <th>realdpi</th>\n",
       "      <th>cpi</th>\n",
       "      <th>m1</th>\n",
       "      <th>tbilrate</th>\n",
       "      <th>unemp</th>\n",
       "      <th>pop</th>\n",
       "      <th>infl</th>\n",
       "      <th>realint</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>1959.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>2710.349</td>\n",
       "      <td>1707.4</td>\n",
       "      <td>286.898</td>\n",
       "      <td>470.045</td>\n",
       "      <td>1886.9</td>\n",
       "      <td>28.98</td>\n",
       "      <td>139.7</td>\n",
       "      <td>2.82</td>\n",
       "      <td>5.8</td>\n",
       "      <td>177.146</td>\n",
       "      <td>0.00</td>\n",
       "      <td>0.00</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>1959.0</td>\n",
       "      <td>2.0</td>\n",
       "      <td>2778.801</td>\n",
       "      <td>1733.7</td>\n",
       "      <td>310.859</td>\n",
       "      <td>481.301</td>\n",
       "      <td>1919.7</td>\n",
       "      <td>29.15</td>\n",
       "      <td>141.7</td>\n",
       "      <td>3.08</td>\n",
       "      <td>5.1</td>\n",
       "      <td>177.830</td>\n",
       "      <td>2.34</td>\n",
       "      <td>0.74</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>1959.0</td>\n",
       "      <td>3.0</td>\n",
       "      <td>2775.488</td>\n",
       "      <td>1751.8</td>\n",
       "      <td>289.226</td>\n",
       "      <td>491.260</td>\n",
       "      <td>1916.4</td>\n",
       "      <td>29.35</td>\n",
       "      <td>140.5</td>\n",
       "      <td>3.82</td>\n",
       "      <td>5.3</td>\n",
       "      <td>178.657</td>\n",
       "      <td>2.74</td>\n",
       "      <td>1.09</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>1959.0</td>\n",
       "      <td>4.0</td>\n",
       "      <td>2785.204</td>\n",
       "      <td>1753.7</td>\n",
       "      <td>299.356</td>\n",
       "      <td>484.052</td>\n",
       "      <td>1931.3</td>\n",
       "      <td>29.37</td>\n",
       "      <td>140.0</td>\n",
       "      <td>4.33</td>\n",
       "      <td>5.6</td>\n",
       "      <td>179.386</td>\n",
       "      <td>0.27</td>\n",
       "      <td>4.06</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>1960.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>2847.699</td>\n",
       "      <td>1770.5</td>\n",
       "      <td>331.722</td>\n",
       "      <td>462.199</td>\n",
       "      <td>1955.5</td>\n",
       "      <td>29.54</td>\n",
       "      <td>139.6</td>\n",
       "      <td>3.50</td>\n",
       "      <td>5.2</td>\n",
       "      <td>180.007</td>\n",
       "      <td>2.31</td>\n",
       "      <td>1.19</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "     year  quarter   realgdp  realcons  realinv  realgovt  realdpi    cpi  \\\n",
       "0  1959.0      1.0  2710.349    1707.4  286.898   470.045   1886.9  28.98   \n",
       "1  1959.0      2.0  2778.801    1733.7  310.859   481.301   1919.7  29.15   \n",
       "2  1959.0      3.0  2775.488    1751.8  289.226   491.260   1916.4  29.35   \n",
       "3  1959.0      4.0  2785.204    1753.7  299.356   484.052   1931.3  29.37   \n",
       "4  1960.0      1.0  2847.699    1770.5  331.722   462.199   1955.5  29.54   \n",
       "\n",
       "      m1  tbilrate  unemp      pop  infl  realint  \n",
       "0  139.7      2.82    5.8  177.146  0.00     0.00  \n",
       "1  141.7      3.08    5.1  177.830  2.34     0.74  \n",
       "2  140.5      3.82    5.3  178.657  2.74     1.09  \n",
       "3  140.0      4.33    5.6  179.386  0.27     4.06  \n",
       "4  139.6      3.50    5.2  180.007  2.31     1.19  "
      ]
     },
     "execution_count": 46,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data = pd.read_csv('../examples/macrodata.csv')\n",
    "data.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 56,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "1959Q1    1959.0\n",
       "1959Q2    1959.0\n",
       "1959Q3    1959.0\n",
       "1959Q4    1959.0\n",
       "1960Q1    1960.0\n",
       "Freq: Q-DEC, Name: year, dtype: float64"
      ]
     },
     "execution_count": 56,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data.year[:5]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 57,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "1959Q1    1.0\n",
       "1959Q2    2.0\n",
       "1959Q3    3.0\n",
       "1959Q4    4.0\n",
       "1960Q1    1.0\n",
       "Freq: Q-DEC, Name: quarter, dtype: float64"
      ]
     },
     "execution_count": 57,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data.quarter[:5]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "通过把这些数组传递给PeriodIndex，并指定频度，我们可以把这些合并得到一个新的DataFrame："
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 53,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "PeriodIndex(['1959Q1', '1959Q2', '1959Q3', '1959Q4', '1960Q1', '1960Q2',\n",
       "             '1960Q3', '1960Q4', '1961Q1', '1961Q2',\n",
       "             ...\n",
       "             '2007Q2', '2007Q3', '2007Q4', '2008Q1', '2008Q2', '2008Q3',\n",
       "             '2008Q4', '2009Q1', '2009Q2', '2009Q3'],\n",
       "            dtype='int64', length=203, freq='Q-DEC')"
      ]
     },
     "execution_count": 53,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "index = pd.PeriodIndex(year=data.year, quarter=data.quarter, \n",
    "                       freq='Q-DEC')\n",
    "index"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 54,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "data.index = index"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 58,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>year</th>\n",
       "      <th>quarter</th>\n",
       "      <th>realgdp</th>\n",
       "      <th>realcons</th>\n",
       "      <th>realinv</th>\n",
       "      <th>realgovt</th>\n",
       "      <th>realdpi</th>\n",
       "      <th>cpi</th>\n",
       "      <th>m1</th>\n",
       "      <th>tbilrate</th>\n",
       "      <th>unemp</th>\n",
       "      <th>pop</th>\n",
       "      <th>infl</th>\n",
       "      <th>realint</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>1959Q1</th>\n",
       "      <td>1959.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>2710.349</td>\n",
       "      <td>1707.4</td>\n",
       "      <td>286.898</td>\n",
       "      <td>470.045</td>\n",
       "      <td>1886.9</td>\n",
       "      <td>28.98</td>\n",
       "      <td>139.7</td>\n",
       "      <td>2.82</td>\n",
       "      <td>5.8</td>\n",
       "      <td>177.146</td>\n",
       "      <td>0.00</td>\n",
       "      <td>0.00</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1959Q2</th>\n",
       "      <td>1959.0</td>\n",
       "      <td>2.0</td>\n",
       "      <td>2778.801</td>\n",
       "      <td>1733.7</td>\n",
       "      <td>310.859</td>\n",
       "      <td>481.301</td>\n",
       "      <td>1919.7</td>\n",
       "      <td>29.15</td>\n",
       "      <td>141.7</td>\n",
       "      <td>3.08</td>\n",
       "      <td>5.1</td>\n",
       "      <td>177.830</td>\n",
       "      <td>2.34</td>\n",
       "      <td>0.74</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1959Q3</th>\n",
       "      <td>1959.0</td>\n",
       "      <td>3.0</td>\n",
       "      <td>2775.488</td>\n",
       "      <td>1751.8</td>\n",
       "      <td>289.226</td>\n",
       "      <td>491.260</td>\n",
       "      <td>1916.4</td>\n",
       "      <td>29.35</td>\n",
       "      <td>140.5</td>\n",
       "      <td>3.82</td>\n",
       "      <td>5.3</td>\n",
       "      <td>178.657</td>\n",
       "      <td>2.74</td>\n",
       "      <td>1.09</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1959Q4</th>\n",
       "      <td>1959.0</td>\n",
       "      <td>4.0</td>\n",
       "      <td>2785.204</td>\n",
       "      <td>1753.7</td>\n",
       "      <td>299.356</td>\n",
       "      <td>484.052</td>\n",
       "      <td>1931.3</td>\n",
       "      <td>29.37</td>\n",
       "      <td>140.0</td>\n",
       "      <td>4.33</td>\n",
       "      <td>5.6</td>\n",
       "      <td>179.386</td>\n",
       "      <td>0.27</td>\n",
       "      <td>4.06</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1960Q1</th>\n",
       "      <td>1960.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>2847.699</td>\n",
       "      <td>1770.5</td>\n",
       "      <td>331.722</td>\n",
       "      <td>462.199</td>\n",
       "      <td>1955.5</td>\n",
       "      <td>29.54</td>\n",
       "      <td>139.6</td>\n",
       "      <td>3.50</td>\n",
       "      <td>5.2</td>\n",
       "      <td>180.007</td>\n",
       "      <td>2.31</td>\n",
       "      <td>1.19</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "          year  quarter   realgdp  realcons  realinv  realgovt  realdpi  \\\n",
       "1959Q1  1959.0      1.0  2710.349    1707.4  286.898   470.045   1886.9   \n",
       "1959Q2  1959.0      2.0  2778.801    1733.7  310.859   481.301   1919.7   \n",
       "1959Q3  1959.0      3.0  2775.488    1751.8  289.226   491.260   1916.4   \n",
       "1959Q4  1959.0      4.0  2785.204    1753.7  299.356   484.052   1931.3   \n",
       "1960Q1  1960.0      1.0  2847.699    1770.5  331.722   462.199   1955.5   \n",
       "\n",
       "          cpi     m1  tbilrate  unemp      pop  infl  realint  \n",
       "1959Q1  28.98  139.7      2.82    5.8  177.146  0.00     0.00  \n",
       "1959Q2  29.15  141.7      3.08    5.1  177.830  2.34     0.74  \n",
       "1959Q3  29.35  140.5      3.82    5.3  178.657  2.74     1.09  \n",
       "1959Q4  29.37  140.0      4.33    5.6  179.386  0.27     4.06  \n",
       "1960Q1  29.54  139.6      3.50    5.2  180.007  2.31     1.19  "
      ]
     },
     "execution_count": 58,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 60,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "1959Q1    0.00\n",
       "1959Q2    2.34\n",
       "1959Q3    2.74\n",
       "1959Q4    0.27\n",
       "1960Q1    2.31\n",
       "Freq: Q-DEC, Name: infl, dtype: float64"
      ]
     },
     "execution_count": 60,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data.infl[:5]"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python [py35]",
   "language": "python",
   "name": "Python [py35]"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.5.2"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 0
}