tds wrote:Is there a way to sum the model weights (for a set of models which contain a specific covariate) automatically in MARK?
Thank you.
Conditionally, yes -- if you build your models using the 'subset models' option (and are using a fairly recent version of MARK). See section 6.5.1 in Chapter 6 of the MARK book, and especially the - sidebar - starting on p. 37.
If that approach doesn't work for you (and using the subset models approach isn't always practical, or desirable), you can do it 'by hand', if you're fairly clever at using some language to extract what you need from the .DBF file. The .DBF file can be sucked into Excel, then exported/saved and an .xls spreadsheet. From there, you need to figure out a way to parse the variables from the model names in the file (in fact, that is the only hard part, or not, depending on your programming skills). Part of why the 'subset models' approach is convenient is that MARK handles model naming, and then summing over models, automatically.