Variance Co-variance Matrix (Eviews8)

[bjohn]

I have estimated an equation using OLS, and I can then view the variance-covariance matrix by clicking on the "View" button and then selecting the "variance-covariance" option.

However, when I try to use this in a programme I come across a problem. Specifically, when I run the line of code

matrix covar_mat = @cholesky(model.@coefcov)

I get the error message saying "Near singular matrix". This is confusing because I can see the variance-covariance matrix myself when I click on the object.

Is there a way to fix this?

[EViews Gareth]

Could you provide the workfile?

[bjohn]

I can provide the workfile- can I upload it somewhere or should I send it via email?

PS: I notice that if I estimate with regular or HAC standard errors the problem doesn't exist, but if I estimate with white standard errors, I get the "near singular matrix" message.

[EViews Gareth]

You can either attach it to a post here, or email it to support@eviews.com

[bjohn]

I have emailed the file.