EViews.com

Hi everyone,

I hope you can help me with this question.

I am working on my master thesis using a dated panel set and I want to create a correlation matrix for my variables included in the regression. However, I cannot find where I conduct this correlation matrix with respect to that my data is panel data. When I read the Eviews help guide, see attached picture, it says that there should be a function called "Panel Covariance" but I cannot find this on my computer (using Eviews 11). I have inserted my data as dated panel of course from start.

Can anyone help me with this? I would be really thankful for any advice you can give!

Regards,
Jesper

Panel Covariance computes the covariance across panels for a single series, and as such is available from the Series View menu.

Hi Gareth,

Thanks for your answer!

I see, so there is no way to calculate correlations between my variables with respect to panel data? In other words that the variables for the same unit (in my case, firm) will have some natural dependence?

Thanks in advance!

You'll need to describe more precisely what you're trying to do mathematically.

Okay, I give it a try.

As far as I know, calculating correlations between variables from a panel data set disregard that for the variables of the same unit (in my case, firm) may depend on its previous value, i.e. that autocorrelation most likely is present. Therefore, I wonder if there is a way to calculate correlations that take this autocorrelation into account?

Is that a more precise description of what I am trying to do?

Regards,
Jesper

It helps a bit, but we are still unclear as to what it is you are trying to do.

What do you mean by accounting for the autocorrelation?

Do you wish to model it and compute the correlations of the residuals? Perhaps compute a long run covariance?

What do you wish to do with the results?

I might just be wrong about that there exists a way to conduct a correlation matrix for panel data with respect to the autocorrelation. I am thankful for your help anyway, now I know that I probably was wrong.

Have a nice day!

Perhaps there is, but to be honest I'm still not clear as to what you want to compute.