Additive effect of time only on some age groups

[scha]

Hi all,

I am still running models with Rmark to investigate senescence on birds and more particularly effect of climate on bird ageing. This time, my problem come from the fact that I need to test an additive effect of time on some age classes survival probabilities (but not on all). I cannot succeed in building the design matrix I need.

I have 6 age groups ([0,1) [1,2) [2,3) [3,4) [4,28) [28,37];I start at age 0) in a variable called ”covar_age”. I wish to test the effect of time only on the four young age classes ([0,1) [1,2) [2,3) [3,4)). It is very easy to test interaction effect between particular age and time by creating an age dummy variable for each age and use :time, but additive effect is quite less easy. By using “+time”, I test an additive effect of time on all ages and I don't have a solution to have an additive effect only on some selected age classes

I have read in an old post (a bit similar) that it was not possible to do that with Rmark (at the date of post). So what about now? Is there a possibility to use formula to do that? Or Am I oblige to use design matrix?

Thank a lot for your reply

[jlaake]

Not sure what post you were reading but this has always been possible. Create a new variable young which is 1 for your young age classes <4 and 0 otherwise. Then use ~age + young:time and it will create an additive time effect for only the young age classes and the other age classes will be constant (no time effect).

--jeff