自动为 AutoRegression 模型 statsmodels 选择滞后 [英] Automatically select lags for AutoRegression model statsmodels

问题描述

在 statsmodels v0.10.1 中，无需在 自回归 AR(p) 模型.如果您选择不指定滞后数，模型会为您选择最适合自动运行模型的一个.在新版本 0.11.1 中，此模型现在称为 AutoReg，现在似乎滞后是强制性的.有没有办法让这个模型在新版本中选择最适合你的滞后数?

In statsmodels v0.10.1 there was no need to choose the number of lags in Autoregressive AR(p) model. If you chose not to specify the number of lags, the model would have chosen the best one for you which was ideal for running the model automatically. In the new version 0.11.1 this model is now called AutoReg, and it seems that the lags are now mandatory. Is there a way to make this model in the new version choose the best number of lags for you?

推荐答案

现在是一个独立的函数，statsmodels.tsa.ar_model.ar_select_order

It is now a standalone function, statsmodels.tsa.ar_model.ar_select_order

https://www.statsmodels.org/devel/generated/statsmodels.tsa.ar_model.ar_select_order.html?highlight=ar_select

这篇关于自动为 AutoRegression 模型 statsmodels 选择滞后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！