add() function and nest survival

[B.K. Sandercock]

I'm using the add() function in the design matrix to model nest survival (pgs 11-21 and 17-14 of the Mark manual). The nest survival data for each nest have an individual covariate (var) that is the age of the nest on the first day of the exposure period. The DM looks like this:
1 add(var, 0)
1 add(var, 1)
1 add(var, 2)
1 add(var, 3)
1 add(var, 4)
1 add(var, 5)
etc.
and the estimates of the real parameters look like this:
95% Confidence Interval
Parameter Estimate Standard Error Lower Upper
------------------------- -------------- -------------- -------------- --------------
1:S 0.9210123 0.0253946 0.8547060 0.9585274
2:S 0.9192959 0.0237393 0.8588373 0.9552110
3:S 0.9175455 0.0221216 0.8625172 0.9517797
4:S 0.9157606 0.0205826 0.8656629 0.9482916
5:S 0.9139407 0.0191790 0.8681611 0.9448340
etc.

If I run the model with mean individual covariate values, it will give real parameter estimates for a mean covariate value of -7 or so because the coding for var is the age of nests on day 1 of the exposure period, where -7 would be seven days before nest initiation. My question is this - what should I set the covariate value to be if I rerun the model with the User-specified Covariate Values? Should the value be set to 1 for day 1 at the start of the nesting cycle? Or should the value be set to the midpoint of the nesting cycle, say a value of 10 for a 20-day incubation period? Thanks for any insights.

I noticed the format of the add() function differs among the examples in different chapters in the Mark manual. It is add(var,0) at pg 17-14, but is add(0,var) at pg 11-21. Will the add() function accept arguments in either format, or is the add() function order-sensitive for the two arguments?

Thanks, Brett.

[gwhite]

Brett:
You should set age to 1 if you want to know the effect of age for days 1-20 of the season. You don't mention whether you have a season effect (T or T^2) in the model. If you do have a season effect, then you may want to do several runs to generate age curves for various times during the season.

add(Var,1) gives identical results to add(1,Var). Same is true of product.

Gary