Variance components problem

[will peach]

Dear Gary/Mark-ers

I have hit a problem with the variance components facility in MARK 4.1 (running on windows 2000) that I cannot resolve (I have tried dummy datasets and asked around).

I am wanting to estimate the proportion of temporal process variance in a set of phi estimates accounted for by various temporal covariates (as advocated by Loison et al. 2002, Euring proceedings). I can estimate process variance from the time-specific real parameter estimates no problem, and I get a plausible looking process variance estimate. The problem comes when I try to define the user-defined design matrix (DM) within the variance components dialog window (I presume I need to use that). I specify real parameters, then list the parameters to include, then click user-defined DM, then OK. I am then asked to enter the number of DM columns, which I do. The DM appears (greyed out), and over the top of that appears an information window that states: 'When you have filled out the DM, click OK to continue'. Problem is I cannot access the DM window. If I click OK (my only option), MARK hangs up (endless egg timer icon) presumably as the DM is undefined.

Am I doing something wrong? Is there a memory problem? Should I re-install MARK?

Any advice much appreciated

Will Peach

[gwhite]

Will:
It sounds like you are doing everything correctly, except I don't understand why you can't access the DM when the message 'When you have filled out the DM, click OK to continue' comes up over the top of it. This works fine on my XP system. Maybe others are having problems with 2000?

Gary