Parameter issue with survival model output

questions concerning analysis/theory using program MARK

Parameter issue with survival model output

Postby rwiltzen » Thu Jan 06, 2022 9:58 pm

Hi everyone,

My first forum post!

I'm using Program Mark and Rmark for analysis of GLMs on nest and brood survival in rock ptarmigan.

My output tables look fine except for the number of parameters, specifically the covariate (time) on survival, or S(~Time), which is the 2nd model from the top. ]

The entire season was 48 days of nest survival, so shouldn't there be 48 parameters instead of 9 parameters in the output table? The text file shows that all 48 occasions are nested in the model. If the output table had 48 parameters listed, this time model would definitely not be my 2nd model.

My question is: why does the model output only show 9 parameters but there should be 48 parameters for total occasions survived?

Also, I can't figure out how to attach images. There's "Img" option above, but it only puts text in my post.

Thanks for your time!
rwiltzen
 
Posts: 1
Joined: Wed Dec 29, 2021 12:53 am

Re: Parameter issue with survival model output

Postby cooch » Fri Jan 07, 2022 9:47 am

rwiltzen wrote:Hi everyone,

My first forum post!

I'm using Program Mark and Rmark for analysis of GLMs on nest and brood survival in rock ptarmigan.

My output tables look fine except for the number of parameters, specifically the covariate (time) on survival, or S(~Time), which is the 2nd model from the top. ]

The entire season was 48 days of nest survival, so shouldn't there be 48 parameters instead of 9 parameters in the output table? The text file shows that all 48 occasions are nested in the model. If the output table had 48 parameters listed, this time model would definitely not be my 2nd model.


I'm guessing you're looking at the beta estimates, rather than the real estimates. In classic MARK, try clicking 'one button to the right'. In fact, if you hover over each of the buttons in the toolbar, you'll see a little 'help balloon' pop up indicating what parameters you're looking at.

In RMark output, the beta and real estimates are presented in that order. Just search for 'Real parameter'.


Also, I can't figure out how to attach images. There's "Img" option above, but it only puts text in my post.

Thanks for your time!


Image embedding is discussed in the FAQ: viewtopic.php?f=29&t=1401

Having said that, 9/10 times you don't need to embed images - simply cut-paste the code (input, or output listing) - discussed in the same FAQ.
cooch
 
Posts: 1628
Joined: Thu May 15, 2003 4:11 pm
Location: Cornell University


Return to analysis help

Who is online

Users browsing this forum: No registered users and 16 guests