如何计算python中的标准化残差? [英] How can I calculate standardized residuals in python?

问题描述

如何根据 arima 模型 sarimax 函数计算标准化残差?

How would I calculated standartized residuals from arima model sarimax function?

假设我们有一些基本模型:

lets say we have some basic model:

import pandas as pd
import numpy as np
import matplotlib.pyplot as plt
import seaborn as sns
sns.set(style='ticks', context='poster')
from statsmodels.tsa.statespace.sarimax import SARIMAX
from statsmodels.tsa.seasonal import seasonal_decompose
import seaborn as sns
#plt.style.use("ggplot")
import pandas_datareader.data as web
import pandas as pd
import statsmodels.api as sm
import scipy
import statsmodels.stats.api as sms
import matplotlib.pyplot as plt
import datetime

model = SARIMAX(df, order = (6, 0, 0), trend = "c");
model_results = model.fit(maxiter = 500);
print(model_results.summary());

我需要标准化器，所以当我们使用 model_results.plot_diagnostics(figsize = (16, 10)); 函数，然后只是基本的 plot 函数残差应该看起来一样.

I need standardizer so when we use model_results.plot_diagnostics(figsize = (16, 10)); function and then just basic plot function residuals should look the same.

推荐答案

我认为你可以使用函数internally_studentized_residual"；来自 https://stackoverflow.com/a/57155553/14294235

I think you can use the function "internally_studentized_residual" from https://stackoverflow.com/a/57155553/14294235

它应该是这样工作的:

model = SARIMAX(df, order = (6, 0, 0), trend = "c");

model_results = model.fit(maxiter = 500);

model_fittebd_y = model_results.fittedvalues

resid_studentized = internally_studentized_residual(df,model_fitted_y)
resid_studentized = -resid_studentized 

plt.plot(resid_studentized)
plt.axhline(y=0, color='b', linestyle='--')
plt.show()

这篇关于如何计算python中的标准化残差?的文章就介绍到这了，希望我们推荐的答案对大家有所帮助，也希望大家多多支持IT屋！