This is probably a silly question ... but I'm stumped so I'm asking it anyway.
I'm using the next survival model to get DSRs for telemetered animals. This is a new model for me in MARK. I have the following input file:
/* 11 */ 1 64 93 1 1 1 2.74 ;
/* 12 */ 1 514 543 1 1 1 3.23 ;
/* 13 */ 1 672 697 1 1 1 4.02 ;
/* 14 */ 1 173 173 0 1 0 2.3 ;
/* 17 */ 1 537 553 1 1 0 2.69 ;
/* 18 */ 1 548 548 0 1 1 3.35 ;
/* 22 */ 1 349 349 0 1 1 2.38 ;
/* 23 */ 1 315 315 0 1 1 4.29 ;
/* 25 */ 1 302 302 0 1 1 3.26 ;
/* 27 */ 1 245 245 0 1 0 2.73 ;
/* 28 */ 1 198 198 0 1 1 2.48 ;
/* 29 */ 1 200 200 0 1 1 4.41 ;
/* 30 */ 1 87 87 0 1 1 3.65 ;
/* 31 */ 1 94 94 0 1 1 3.22 ;
The comment being an index number for each individual. I then have the 5 requisite columns followed by 2 individual covariates.
I am having two problems: First, when I create a new MARK project it warns me that the file has only 14 semicolons and ask me verify that the file is a valid mark input file; Second, and more importantly, when I try to run a basic intercept only model in MARK, I get the error that the input must contain only 1's and 0's.
What gives? Obviously the file contains only 14 semicolon as it's only 14 lines long. I also understand that MARK creates a true encounter history from the unusual input required for this model ... but I don't understand why it would be generating anything that would be invalid input. I assumed that the number of encounter occasions should be equal to the largest number I have in the third column as it represents the last day any "nest" was checked. I couldn't find this explicitly in the manual, but it made sense to me. I selected 1 group, and 2 covariates in the setup screen (and named the covariates). I also set survival to constant (though I'm not sure how that would make a difference).
Please, I'm open to suggestions.
Matt