EViews.com

Hi,

Not sure if this is a bug, but I am experiencing some rather strange behaviour with my model. A bunch of variables are made exogenous/overridden and then the model solves for the the add-factors. These add-factors
are extracted.

A new scenario is then created with the add-factors made exogenous (using exclude/override). The model solves and the variables with the add-factors match their exogenous counterparts. So the model is doing as intended. I can re-solve the model any number of times and the result is the same.

I then save the model to disk. When I re-open the model and solve the exact same scenario I get an error ( a numerical error, indicating the model is going off-course, i.e. log of a non-positive number).

I should stress, all that is happening here is that I am saving a workfile with a model that solves to disk, re-opening and trying to solve the model (using the exact same solve options, sample and scenario) and it does not solve. This leads me to think that something may be cached in memory that gets cleared when the workfile closes? But that is just a guess.

Any thoughts?

Thanks

Adam

Hello,

That certainly sounds like a bug, but I have not been able to recreate the issue. Could you post an EViews program/workfile that demonstrates the problem?

Thanks Matt, unfortunately I can't share the workfile / program. But I will try to re-create an example that I can share.