现有 Dataframe 的动态列和 python 中先前记录的赋值

Question

我正在尝试动态地将新列插入到我现有的数据框中，并从当前索引 5、索引 10、索引 15、索引 20----索引 300 的先前记录中分配值。基本上我需要将 180 列（之前记录中的 3 列）及其值添加到数据框的每一行。

现有 DF 输出：

index----Price----Bid------ask

1        70.15    70.10    70.20
2        70.18    70.13    70.16
3    70.09    70.09    70.10
4    70.14    70.11    70.14 
5    70.13    70.12    70.13
6    70.16    70.16    70.19
7    70.14    70.12    70.16
:
:
n

新 DF 应如下所示：-

index----Price----Bid-----Ask----5_Price--5_Bid--5_Ask--10_Price--10_Bid--10_Ask.....300_Price--300_Bid--300_Ask
1        70.15    70.10    70.20  70.03   70.04  70.05  70.05     70.06   70.07      70.14       70.14    70.16
2        70.18    70.13    70.16  70.01   70.02  70.03  70.09     70.09   70.10      70.17       70.16    70.17
3    70.09    70.09    70.10  70.05   70.03  70.05  70.06     70.04   70.06      70.20       70.18    70.19
4    70.14    70.11    70.14  70.07   70.09  70.10  70.05     70.06   70.07      70.14       70.14    70.16
5    70.13    70.12    70.13  70.03   70.04  70.05  70.05     70.06   70.07      70.14       70.14    70.16
6    70.16    70.16    70.19  70.18   70.16  70.19  70.13     70.11   70.12      70.17       70.18    70.19
7    70.14    70.12    70.16  70.20   70.21  70.19  70.21     70.19   70.21      70.08       70.07    70.09
:
:
n

代码：有效，但即使是 50000 条记录（几个小时）也需要很长时间

t = 5
p = 305

last = '_Last'
cbid = '_Cbid'
cask = '_Cask'

for i in range(p, df.shape[0]):
    n = i - 300
    for j in range(5,p,5):
        in_last = str(j) + last
        in_cbid = str(j) + cbid
        in_cask = str(j) + cask
        df.loc[i, in_last] = df.loc[n+j, "Nusdinr Last"]
        df.loc[i, in_cbid] = df.loc[n+j, "Nusdinr Close Bid"]
        df.loc[i, in_cask] = df.loc[n+j, "Nusdinr Close Ask"]...

Answer 1

这里有一个更有效的方法，使用 'shift'。

for shift_days in range(0, 300, 5):
    col_names = [f"{col}_{shift_days}" for col in ["Price", "Bid", "ask"]]
    df[col_names] = df[["Price", "Bid", "ask"]].shift(shift_days)

现有 Dataframe 的动态列和 python 中先前记录的赋值

Dynamic columns to existing Dataframe and value assignment from previous record in python

python

dynamic-columns

dataframe

现有 DF 输出：

新 DF 应如下所示：-