如何改进 LAG 功能？

Question

我有以下 table:

time	comsales	indsales
1	20.96	127.3
2	21.4	130
3	21.96	132.7
4	21.52	129.4
5	22.39	135
6	22.76	137.1
7	23.48	141.2
8	23.66	142.8
9	24.1	145.5
10	24.01	145.3
11	24.54	148.3
12	24.3	146.4
13	25	150.2
14	25.64	153.1
15	26.36	157.3
16	26.98	160.7
17	27.52	164.2
18	27.78	165.6
19	28.24	168.7
20	28.78	171.7

我正在尝试使用以下代码在 python 中实现 LAG 函数

import pandas as pd
import numpy as np
import matplotlib.pyplot as plt
import statsmodels.api as sm

dataset = pd.read_csv('Data/blaisdellcomp.csv', index_col=0)
y = dataset['comsales']
x = dataset['indsales']

X = sm.add_constant(x)
model = sm.OLS(y, X).fit()
print(model.summary())

residuals = model.resid
residuals = residuals.reset_index(drop=True)

lagged_residuals = residuals.rolling(1).sum()
lagged_residuals.drop(lagged_residuals.tail(1).index,inplace=True)
lagged_residuals.loc[-1] = lagged_residuals[0]  # adding a row
lagged_residuals.index = lagged_residuals.index + 1  # shifting index
lagged_residuals.sort_index(inplace=True)
lagged_residuals = lagged_residuals.reset_index(drop=True)

为了能够在残差中产生滞后，我使用了很多代码来实现滞后。有没有更好的方法来实现残差的滞后？

Answer 1

您似乎在寻找 Pandas shift 方法：

lagged_residuals = residuals.shift(1)

如何改进 LAG 功能？

How can I improve the LAG function?

python-3.x

statsmodels