简体繁体 English

ACF PACF 测定 ARIMA

[英]ACF PACF Determination ARIMA

原文 2021-01-09 14:53:26 5 1 r/ time-series/ arima

Please help me confirm my understanding.请帮助我确认我的理解。 For below graphs, I believe对于下面的图表，我相信

AR(p) = 0 and MA(q) = 0 AR(p) = 0 和 MA(q) = 0

Is that correct?那是对的吗？

1 个解决方案

First, let's learn more...首先，让我们了解更多...

As Aashiq Reza brought the description link, I think the ACF and PACF plots that you shared is like an MA(2) process.由于Aashiq Reza带来了描述链接，我认为您分享的ACF和PACF图就像一个MA(2)过程。 ARIMA(p, i,q) has three elements, p is for AR , i is for difference, and q stands for MA process lag. ARIMA(p, i,q)具有三个元素， p代表AR ， i代表差异， q代表MA过程滞后。 Because the lag parameter defines the lag in the model's regression formula, if both of p and q be zero, then the model is not ARIMA anymore.因为滞后参数定义了模型回归公式中的滞后，如果p和q都为零，则 model 不再是ARIMA 。

My suggestion: probabilistic model selection...我的建议：概率 model 选择...

You can evaluate the correctness of a model for a time-series object using information criteria like AIC and BIC .您可以使用AIC和BIC等information criteria评估时间序列 object 的 model 的正确性。 For example, you have a preset of possible p and q , then you can test each one and get the criteria for it.例如，您有一个可能的p和q的预设，然后您可以测试每一个并获得它的标准。 The model with the least criterion is the best one.标准最少的 model 是最好的。 This link helps with the calculation in python. 此链接有助于 python 中的计算。