《利用python做数据分析》第十章：时间序列分析

import pandas as pd
import numpy as np
import matplotlib.pyplot as plt
%matplotlib inline

//anaconda/lib/python2.7/site-packages/matplotlib/font_manager.py:273: UserWarning: Matplotlib is building the font cache using fc-list. This may take a moment. warnings.warn(‘Matplotlib is building the font cache using fc-list. This may take a moment.’)

from pandas import Series,DataFrame

#### Time Seiries Analysis **** > build-in package time datetime calendar

from datetime import datetime

now = datetime.now()

now

datetime.datetime(2016, 2, 1, 11, 11, 8, 934671) > ** display time right now **

now.year,now.month,now.day

(2016, 2, 1) datetime以毫秒形势存储��和⌚️，**datetime.datedelta**表示两个datetime对象之间的时间差

delta = datetime(2011,1,7) - datetime(2008,6,24,8,15)

显示的前一个是天数，后一个是秒钟 —- delta.days delta.seconds

delta

datetime.timedelta(926, 56700) ### 可以给datetime对象加上或者减去一个或者多个timedelta，会产生一个新对象

from datetime import timedelta

start = datetime(2011, 1, 7)

start + timedelta(12)

datetime.datetime(2011, 1, 19, 0, 0)

start - timedelta(12) * 4

datetime.datetime(2010, 11, 20, 0, 0) > 可见timedelta是以天为单位 #### datetime模块中的数据类型 —– - date | 以公历形式存储日历日期（年、月、日） - time | 将时间存储为时、分、秒、毫秒 - datetime | 存储时间和日期 - timedelta| 比阿诗两个datetime值之间的差（日， 秒， 毫秒） ## str transformed to datetime use ** str ** or ** strftime(invoke a formed str) ** ,datetime object and pandas.Timestamp can be formulated to string

stamp = datetime(2011, 1, 3)

str(stamp)

‘2011-01-03 00:00:00’

stamp.strftime('%Y-%m-%d')

‘2011-01-03’

stamp.strftime('%Y-%m')

‘2011-01’

value = '2011-01-03'

datetime.strptime(value, '%Y-%m-%d')

datetime.datetime(2011, 1, 3, 0, 0)

datestrs = ['7/6/2011','8/6/2011']

[datetime.strptime(x, '%m/%d/%Y') for x in datestrs]

[datetime.datetime(2011, 7, 6, 0, 0), datetime.datetime(2011, 8, 6, 0, 0)] datetime.striptime 是通过已知格式进行日期解析的最佳方式，但每次都要编写格式定义 - 使用dateutil中的parser.parse来实现

from dateutil.parser import parse

parse('2011-01-03')

datetime.datetime(2011, 1, 3, 0, 0) parse的解析能力很强，几乎可以解析一切格式

parse('Jan 31,1997 10:45 PM')

datetime.datetime(1997, 1, 31, 22, 45)

parse('6/30/2011', dayfirst=True)

datetime.datetime(2011, 6, 30, 0, 0)

datestrs

[‘7/6/2011’, ‘8/6/2011’] # pd.to_datetime()

pd.to_datetime(datestrs)

DatetimeIndex([‘2011-07-06’, ‘2011-08-06’], dtype=’datetime64[ns]’, freq=None)

dates = [datetime(2011, 1, 2),datetime(2011,1,5),datetime(2011,1,7),
        datetime(2011,1,8),datetime(2011,1,10),datetime(2011,1,12)]

ts = Series(np.random.randn(6), index=dates)

ts

2011-01-02 0.573974 2011-01-05 -0.337112 2011-01-07 -1.650845 2011-01-08 0.450012 2011-01-10 -1.253801 2011-01-12 -0.402997 dtype: float64

type(ts)

pandas.core.series.Series

ts.index

DatetimeIndex([‘2011-01-02’, ‘2011-01-05’, ‘2011-01-07’, ‘2011-01-08’, ‘2011-01-10’, ‘2011-01-12’], dtype=’datetime64[ns]’, freq=None)

ts + ts[::2]

2011-01-02 1.147949 2011-01-05 NaN 2011-01-07 -3.301690 2011-01-08 NaN 2011-01-10 -2.507602 2011-01-12 NaN dtype: float64

ts[::2]

2011-01-02 0.573974 2011-01-07 -1.650845 2011-01-10 -1.253801 dtype: float64 ## 索引、选取、子集构造

ts['1/10/2011']

-1.2538008746706757 传入可以解释为日期的字符，就可以代替索引

ts['20110110']

-1.2538008746706757

longer_ts=Series(np.random.randn(1000),index=pd.date_range('20000101',periods=1000))

longer_ts

2000-01-01 -1.025498 2000-01-02 -0.913267 2000-01-03 0.240895 2000-01-04 -1.475368 2000-01-05 -1.675558 2000-01-06 1.020005 2000-01-07 0.638097 2000-01-08 0.503482 2000-01-09 -0.541771 2000-01-10 -1.107036 2000-01-11 0.797612 2000-01-12 1.691745 2000-01-13 1.889323 2000-01-14 -0.852126 2000-01-15 -0.987578 2000-01-16 0.558084 2000-01-17 -0.842907 2000-01-18 1.932399 2000-01-19 -1.126650 2000-01-20 -0.529707 2000-01-21 0.116756 2000-01-22 -0.012790 2000-01-23 0.501330 2000-01-24 0.346976 2000-01-25 -0.880443 2000-01-26