limit to number of parameters in newest MARK build?

[Res]

Hello,

I just installed the latest MARK 4.1 build (3 Dec). This version doesn't seem to accept more than 999 parameters in the PIMs. The boxes apparently can only hold numbers with up to 3 digits. In older versions, I was able to use 4-digit numbers. Can I somehow get this version of MARK to accept larger numbers? Or is it possible to make the boxes wider again in future versions, Gary?

It's not that I want to estimate >1000 parameters. But I need to build a flexible underlying structure before I can specify the interesting models in the design matrix.

I also enountered limits of the input box for the Variance-Components estimation. MARK didn't accept more parameters than I could fit into the input box. Is it possible to get MARK to accept more input than acutally fits into the boxes on the screen? Or is this a problem of my computer or Windows (2000), maybe?

Thanks for your help.
Res

[cooch]

Hello,

I just installed the latest MARK 4.1 build (3 Dec). This version doesn't seem to accept more than 999 parameters in the PIMs. The boxes apparently can only hold numbers with up to 3 digits. In older versions, I was able to use 4-digit numbers. Can I somehow get this version of MARK to accept larger numbers? Or is it possible to make the boxes wider again in future versions, Gary?

I'm confused. For a given PIM, there is no way I can conceive of for needing >999 parameters. For a time-dependent PIM, that would imply >999 occasions (unlikely). And even for a model with (say) 30 occasions, 3 age classes, and full-time-dependence, you're still <100 parameters.

The only way I can see getting ahuge number of parameteers is a full agextimexcohort model, which isn't fully idnetifiable anyway.

Again, we're talking about the number of parameters in a PIM, right?