Lukacs young survival & temporal environmental covariates

Forum for discussion of general questions related to study design and/or analysis of existing data - software neutral.

Lukacs young survival & temporal environmental covariates

Postby ursakoce » Tue Nov 21, 2017 11:34 am

Helo!

I am looking for an advice about how to incorporate time dependent environmental covariates in the 'Lukacs young survival from marked adults' model structure. In the MARK tutorial, chapter 19, one is directed to chapter 6 to read the datails about how to include environmental covariates in the design matrix but there I do not find information that would really help me to model time dependent environmental data such as daily precipitation or temperature in Lukacs young survival module ... Our dataset consists of encounter histories of N plover families collected over three years; there are 28 encounter occasions (corresponding to 28-day post-hatching period of the species), three (year) groups, and several individual covariates. Besides the effects of the individual covariates on the apparent survival I would also like to model the effects of three time dependent environmental variables: daily precipitation, maximum daily temperature and maximum daily river flow. I specified time dependent apparent survival parameters in the PIM, which gives 27 parameters for each year group (=81 Phi parameters altogether). (I keep p constant.) But given that the encounter occasions of individual plover families cover different time periods, I cannot simply enter 3x27 daily values into the design matrix to represent the environmental covariate of interest (because these would be different for different plover families). My approach was to calculate average daily values for each encounter occasion accross all plover families from the same year but it seems to me that doing so a lot of information in the data is lost.

I guess I must be missing some bit of information to specify my model correctly and in case these things have been worked out, I would very much appreciate a solution that maybe I simply don't see.

Thank you,
Urša
ursakoce
 
Posts: 2
Joined: Mon Aug 28, 2017 4:30 am

Return to analysis & design questions

Who is online

Users browsing this forum: No registered users and 16 guests