Heterogeneity models

questions concerning analysis/theory using program MARK

Re: Heterogeneity models

Postby WiPhi » Wed Apr 19, 2017 7:05 pm

Found this post and I think it relates to a fundamental analysis question that I would hope someone can verify. When defining Pledger mixtures, the mixtures occur WITHIN each group (e.g. sex or ageclass or sexXageclass) however groups have been defined (rather than groups occur within mixtures). The assumption then is that the level of mixture (estimates for pi) are the same within each group, i.e., the same proportion of low encounter individuals and high encounter individuals in each ageclass.
WiPhi
 
Posts: 24
Joined: Thu Mar 31, 2016 6:15 pm

Re: Heterogeneity models

Postby cooch » Wed Apr 19, 2017 7:54 pm

WiPhi wrote:Found this post and I think it relates to a fundamental analysis question that I would hope someone can verify. When defining Pledger mixtures, the mixtures occur WITHIN each group (e.g. sex or ageclass or sexXageclass) however groups have been defined (rather than groups occur within mixtures). The assumption then is that the level of mixture (estimates for pi) are the same within each group, i.e., the same proportion of low encounter individuals and high encounter individuals in each ageclass.


If I'm interpreting your commnent/question, the answer is 'no'. Each group (if specified using group attributes) gets its own mixture -- meanining, the probability of assignment to a given mixture group is unique to that group (Unless you constrain them to be the same in a reduced parameter model). For example, take the Diper data set -- two sexes, male and female. Each sex class has its own $\pi$ parameter.

Note -- $\pi$ is a probability, and should not, generally, be interpreted as a proportion. Please see section on 'interpeting' $\pi$ in Chapter 14 (currently on p. 27).
cooch
 
Posts: 1240
Joined: Thu May 15, 2003 4:11 pm
Location: Cornell University

Re: Heterogeneity models

Postby WiPhi » Wed Apr 26, 2017 2:42 pm

Yes, thanks, that response directly addressed my question. I was interested in understanding what is going on in the reduced parameter case (I am using R-mark, considering a model with pi formulated as a function of a mixture parameter (h-1 cases) and as a function of the mixture and group parameter (with groups defined by sex and age).
WiPhi
 
Posts: 24
Joined: Thu Mar 31, 2016 6:15 pm

Previous

Return to analysis help

Who is online

Users browsing this forum: No registered users and 1 guest

cron