python笔记：3.2.2.8pandas数据操作_时间序列之频率转换及重采样

# -*- coding: utf-8 -*-
"""
Created on Fri May 24 19:59:16 2019

@author: User
"""

import pandas as pd
import numpy as np


print("\n 按工作日生成的索引：")
date=pd.date_range('2019-03-13','2019-03-29',freq='B')
sample=pd.DataFrame({'open_p':np.random.randn(13),'close_p':np.random.randn(13)},
                 index=date)
print(sample)

print("\n 按日历日生成的索引：")
print(sample.asfreq(freq='D'))

print("\n sample.resample('12H').ffill() 按照半天频率进行上采样，指定插值方式：")
print(sample.resample('12H').ffill())

print("\n sample.resample('4D').ohlc() 按照4天频率进行采样(起点 最大 最小 终点数据)：")
print(sample.resample('4D').ohlc())

print("\n sample.resample('4D').ohlc() 按照4天频率进行采样(起点 最大 最小 终点数据)：")
print(sample.groupby(lambda x:x.week).mean())

print("\n sample[['open_p','close_p']].groupby(lambda x:x.month).mean() 提取股票月开收盘价：")
print(sample[['open_p','close_p']].groupby(lambda x:x.month).mean())

运行：

 按工作日生成的索引：
              open_p   close_p
2019-03-13  1.237511  0.167589
2019-03-14 -2.011978 -0.575576
2019-03-15  0.548313  0.401648
2019-03-18 -0.981720  0.173023
2019-03-19 -0.532745 -0.413077
2019-03-20  0.898792 -0.029455
2019-03-21 -0.559489 -0.472909
2019-03-22 -0.899932 -0.304375
2019-03-25 -1.850232  0.436640
2019-03-26  0.306646  0.518828
2019-03-27  1.242612 -1.255874
2019-03-28  0.616150  1.920751
2019-03-29  0.043343  1.076985

 按日历日生成的索引：
              open_p   close_p
2019-03-13  1.237511  0.167589
2019-03-14 -2.011978 -0.575576
2019-03-15  0.548313  0.401648
2019-03-16       NaN       NaN
2019-03-17       NaN       NaN
2019-03-18 -0.981720  0.173023
2019-03-19 -0.532745 -0.413077
2019-03-20  0.898792 -0.029455
2019-03-21 -0.559489 -0.472909
2019-03-22 -0.899932 -0.304375
2019-03-23       NaN       NaN
2019-03-24       NaN       NaN
2019-03-25 -1.850232  0.436640
2019-03-26  0.306646  0.518828
2019-03-27  1.242612 -1.255874
2019-03-28  0.616150  1.920751
2019-03-29  0.043343  1.076985

 sample.resample('12H').ffill() 按照半天频率进行上采样，指定插值方式：
                       open_p   close_p
2019-03-13 00:00:00  1.237511  0.167589
2019-03-13 12:00:00  1.237511  0.167589
2019-03-14 00:00:00 -2.011978 -0.575576
2019-03-14 12:00:00 -2.011978 -0.575576
2019-03-15 00:00:00  0.548313  0.401648
2019-03-15 12:00:00  0.548313  0.401648
2019-03-16 00:00:00  0.548313  0.401648
2019-03-16 12:00:00  0.548313  0.401648
2019-03-17 00:00:00  0.548313  0.401648
2019-03-17 12:00:00  0.548313  0.401648
2019-03-18 00:00:00 -0.981720  0.173023
2019-03-18 12:00:00 -0.981720  0.173023
2019-03-19 00:00:00 -0.532745 -0.413077
2019-03-19 12:00:00 -0.532745 -0.413077
2019-03-20 00:00:00  0.898792 -0.029455
2019-03-20 12:00:00  0.898792 -0.029455
2019-03-21 00:00:00 -0.559489 -0.472909
2019-03-21 12:00:00 -0.559489 -0.472909
2019-03-22 00:00:00 -0.899932 -0.304375
2019-03-22 12:00:00 -0.899932 -0.304375
2019-03-23 00:00:00 -0.899932 -0.304375
2019-03-23 12:00:00 -0.899932 -0.304375
2019-03-24 00:00:00 -0.899932 -0.304375
2019-03-24 12:00:00 -0.899932 -0.304375
2019-03-25 00:00:00 -1.850232  0.436640
2019-03-25 12:00:00 -1.850232  0.436640
2019-03-26 00:00:00  0.306646  0.518828
2019-03-26 12:00:00  0.306646  0.518828
2019-03-27 00:00:00  1.242612 -1.255874
2019-03-27 12:00:00  1.242612 -1.255874
2019-03-28 00:00:00  0.616150  1.920751
2019-03-28 12:00:00  0.616150  1.920751
2019-03-29 00:00:00  0.043343  1.076985

 sample.resample('4D').ohlc() 按照4天频率进行采样(起点 最大 最小 终点数据)：
              open_p                      ...   close_p                    
                open      high       low  ...      high       low     close
2019-03-13  1.237511  1.237511 -2.011978  ...  0.401648 -0.575576  0.401648
2019-03-17 -0.981720  0.898792 -0.981720  ...  0.173023 -0.413077 -0.029455
2019-03-21 -0.559489 -0.559489 -0.899932  ... -0.304375 -0.472909 -0.304375
2019-03-25 -1.850232  1.242612 -1.850232  ...  1.920751 -1.255874  1.920751
2019-03-29  0.043343  0.043343  0.043343  ...  1.076985  1.076985  1.076985

[5 rows x 8 columns]

 sample.resample('4D').ohlc() 按照4天频率进行采样(起点 最大 最小 终点数据)：
      open_p   close_p
11 -0.075385 -0.002113
12 -0.415019 -0.209358
13  0.071704  0.539466

 sample[['open_p','close_p']].groupby(lambda x:x.month).mean() 提取股票月开收盘价：
     open_p   close_p
3 -0.149441  0.126477