获取前一天的数据以计算新的列数据

Question

我有以下数据：

High    Low Open    Close   Volume  Adj Close   bcc
Date                            
2018-01-02  2695.889893 2682.360107 2683.729980 2695.810059 3367250000  2695.810059 False
2018-01-03  2714.370117 2697.770020 2697.850098 2713.060059 3538660000  2713.060059 False
2018-01-04  2729.290039 2719.070068 2719.310059 2723.989990 3695260000  2723.989990 False
2018-01-05  2743.449951 2727.919922 2731.330078 2743.149902 3236620000  2743.149902 False
2018-01-08  2748.510010 2737.600098 2742.669922 2747.709961 3242650000  2747.709961 True
... ... ... ... ... ... ... ...
2020-04-13  2782.459961 2721.169922 2782.459961 2761.629883 5274310000  2761.629883 False
2020-04-14  2851.850098 2805.100098 2805.100098 2846.060059 5567400000  2846.060059 False
2020-04-15  2801.879883 2761.540039 2795.639893 2783.360107 5203390000  2783.360107 False
2020-04-16  2806.510010 2764.320068 2799.340088 2799.550049 5179990000  2799.550049 False
2020-04-17  2879.219971 2830.879883 2842.429932 2874.560059 5792140000  2874.560059 False
577 rows × 7 columns

我需要创建一个名为"pct"的列，在其中我得到前一天的close，并将其除以今天的Low。我怎样才能做到这一点？

Answer 1

如果需要，用close列中的前几天除以low列中的实际天数：

df['prev'] = df['Close'].shift(freq='1d').div(df['Low'])
print (df)
High          Low         Open        Close      Volume  
Date                                                                         
2018-01-02  2695.889893  2682.360107  2683.729980  2695.810059  3367250000   
2018-01-03  2714.370117  2697.770020  2697.850098  2713.060059  3538660000   
2018-01-04  2729.290039  2719.070068  2719.310059  2723.989990  3695260000   
2018-01-05  2743.449951  2727.919922  2731.330078  2743.149902  3236620000   
2018-01-08  2748.510010  2737.600098  2742.669922  2747.709961  3242650000   
2020-04-13  2782.459961  2721.169922  2782.459961  2761.629883  5274310000   
2020-04-14  2851.850098  2805.100098  2805.100098  2846.060059  5567400000   
2020-04-15  2801.879883  2761.540039  2795.639893  2783.360107  5203390000   
2020-04-16  2806.510010  2764.320068  2799.340088  2799.550049  5179990000   
2020-04-17  2879.219971  2830.879883  2842.429932  2874.560059  5792140000   
Adj Close    bcc      prev  
Date                                      
2018-01-02  2695.810059  False       NaN  
2018-01-03  2713.060059  False  0.999273  
2018-01-04  2723.989990  False  0.997790  
2018-01-05  2743.149902  False  0.998559  
2018-01-08  2747.709961   True       NaN  
2020-04-13  2761.629883  False       NaN  
2020-04-14  2846.060059  False  0.984503  
2020-04-15  2783.360107  False  1.030606  
2020-04-16  2799.550049  False  1.006888  
2020-04-17  2874.560059  False  0.988933  

如果需要除以今天Low:

#last datetime changed to today
print (df)
High          Low         Open        Close      Volume  
Date                                                                         
2018-01-02  2695.889893  2682.360107  2683.729980  2695.810059  3367250000   
2018-01-03  2714.370117  2697.770020  2697.850098  2713.060059  3538660000   
2018-01-04  2729.290039  2719.070068  2719.310059  2723.989990  3695260000   
2018-01-05  2743.449951  2727.919922  2731.330078  2743.149902  3236620000   
2018-01-08  2748.510010  2737.600098  2742.669922  2747.709961  3242650000   
2020-04-13  2782.459961  2721.169922  2782.459961  2761.629883  5274310000   
2020-04-14  2851.850098  2805.100098  2805.100098  2846.060059  5567400000   
2020-04-15  2801.879883  2761.540039  2795.639893  2783.360107  5203390000   
2020-04-16  2806.510010  2764.320068  2799.340088  2799.550049  5179990000   
2020-04-21  2879.219971  2830.879883  2842.429932  2874.560059  5792140000   
Adj Close    bcc  
Date                            
2018-01-02  2695.810059  False  
2018-01-03  2713.060059  False  
2018-01-04  2723.989990  False  
2018-01-05  2743.149902  False  
2018-01-08  2747.709961   True  
2020-04-13  2761.629883  False  
2020-04-14  2846.060059  False  
2020-04-15  2783.360107  False  
2020-04-16  2799.550049  False  
2020-04-21  2874.560059  False  

---

today = pd.Timestamp.today().floor('D')
print (today)
2020-04-21 00:00:00
print (df.loc[today, 'Low'])
2830.8798829999996
df['prev'] = df['Close'].shift(freq='1d').div(df.loc[today, 'Low'])
print (df)
High          Low         Open        Close      Volume  
Date                                                                         
2018-01-02  2695.889893  2682.360107  2683.729980  2695.810059  3367250000   
2018-01-03  2714.370117  2697.770020  2697.850098  2713.060059  3538660000   
2018-01-04  2729.290039  2719.070068  2719.310059  2723.989990  3695260000   
2018-01-05  2743.449951  2727.919922  2731.330078  2743.149902  3236620000   
2018-01-08  2748.510010  2737.600098  2742.669922  2747.709961  3242650000   
2020-04-13  2782.459961  2721.169922  2782.459961  2761.629883  5274310000   
2020-04-14  2851.850098  2805.100098  2805.100098  2846.060059  5567400000   
2020-04-15  2801.879883  2761.540039  2795.639893  2783.360107  5203390000   
2020-04-16  2806.510010  2764.320068  2799.340088  2799.550049  5179990000   
2020-04-21  2879.219971  2830.879883  2842.429932  2874.560059  5792140000   
Adj Close    bcc      prev  
Date                                      
2018-01-02  2695.810059  False       NaN  
2018-01-03  2713.060059  False  0.952287  
2018-01-04  2723.989990  False  0.958380  
2018-01-05  2743.149902  False  0.962241  
2018-01-08  2747.709961   True       NaN  
2020-04-13  2761.629883  False       NaN  
2020-04-14  2846.060059  False  0.975538  
2020-04-15  2783.360107  False  1.005362  
2020-04-16  2799.550049  False  0.983214  
2020-04-21  2874.560059  False       NaN