Test for multi-collinearity in panel data R

I am running a panel data regression using the package plmin Rand want to control the multi-collinearity between the explanatory variables.
I know that carthere is a function in the package vif(), however, as far as I know, it cannot handle the output of panel data.
plmcan perform other diagnostic operations, such as a single root test, but I have not found a method for calculating multicollinearity.

Is there a way to calculate a similar test for vif, or can I just treat each variable as a time series, leaving the panel information and running the tests using the package car?

I can’t disclose the data, but the problem should be relevant for all panel data models.
The dimension is approximately 1000 observations, more than 50 time periods.
The code I use looks like this:

pdata <- plm.data(RegData, index=c("id","time"))
fixed <- plm(Y~X, data=pdata, model="within")

and then

vif(fixed) 

returns an error.


Thank you in advance.

+4
source share
1 answer

This question was asked with reference to other statistical packages, such as SAS https://communities.sas.com/thread/47675 and Stata http://www.stata.com/statalist/archive/2005-08/msg00018.html , and the general answer was to use a unified model to get VIF. The logic is that since multicollinearity is only an independent variable, there is no need to control individual effects using panel methods.

+9

All Articles