Question

我在pandas数据框中有一些时间序列数据：

 prices.head()
                 Time   A        B      C      D
0 2012-01-02 08:00:30 NaN  47.1650  31.51  58.16
1 2012-01-02 08:01:00 NaN  47.2400  31.48  58.19
2 2012-01-02 08:01:30 NaN  47.2750  31.46  58.21
3 2012-01-02 08:02:00 NaN  47.3250  31.40  58.17
4 2012-01-02 08:02:30 NaN  47.3325  31.42  58.07

我想创建4个包含每天收盘价的新列。我怎么能这样做？

与第1天相关的样本应具有第1天的收盘价，依此类推......

Answer 1

您可以groupby约会，然后选择每个小组的最后一个，然后加入。

df['date'] = df.Time.dt.date
print df.join(df.groupby('date')[['A','B','C','D']].last(), rsuffix='_close', on='date')

                 Time   A        B      C      D        date  A_close  \
0 2012-01-02 08:00:30 NaN  47.1650  31.51  58.16  2012-01-02      NaN   
1 2012-01-02 08:01:00 NaN  47.2400  31.48  58.19  2012-01-02      NaN   
2 2012-01-02 08:01:30 NaN  47.2750  31.46  58.21  2012-01-02      NaN   
3 2012-01-02 08:02:00 NaN  47.3250  31.40  58.17  2012-01-02      NaN   
4 2012-01-02 08:02:30 NaN  47.3325  31.42  58.07  2012-01-02      NaN   

   B_close  C_close  D_close  
0  47.3325    31.42    58.07  
1  47.3325    31.42    58.07  
2  47.3325    31.42    58.07  
3  47.3325    31.42    58.07  
4  47.3325    31.42    58.07

大熊猫从时间序列数据中获得收盘价

1 个答案: