Difference between 3 RMark models

[berg1546]

Hi all,

A friend of mine is starting to work with MARK and RMark and asked me the following question. I wasn't sure how to answer, so I'm posting it for him here:

"I am looking for a good explanation of the differences between the RMark models Robust, RDHet and RDFullHet? I know what the parameters are in each, I am just not sure how the different models estimate the different parameters.

I am getting "better" results from my data with RDHet, but I am not sure why, so hopefully a better understanding of the models will help."

Thanks in advance,
Serge

[cooch]

Hi all,

A friend of mine is starting to work with MARK and RMark and asked me the following question. I wasn't sure how to answer, so I'm posting it for him here:

"I am looking for a good explanation of the differences between the RMark models Robust, RDHet and RDFullHet? I know what the parameters are in each, I am just not sure how the different models estimate the different parameters.

I am getting "better" results from my data with RDHet, but I am not sure why, so hopefully a better understanding of the models will help."

Thanks in advance,
Serge

1\ turn on regular 'classic' MARK, and access 'Help | Data Types'. Look for the models you've mentioned.

2\ you'll see that the models are all RD, and differ in only how the closed population samples are modeled.Some of the models have the finite mixture model to handle heterogeneity, some don't.

3\ to fully understand the differences, you need to have a thorough read of Chapter 14 in the MARK book. It covers the basic distinctions among models with and without heterogeneity. There isn't much discussion of these at all in the RD chapter (15), since at that point, the assumption is that you've read and absorbed Chapter 14 first.

So you (or whoever you're posting for) needs to read 14 + 15. And, for future, 'better results' means little of anything useful without qualification. Better, by what criterion?