RELEASE GOF Lentgth of command

[gabrielcaetano]

When running RELEASE GOF test for my CJS model, i get no results in the output file, instead i get this:


INPUT --- proc title tropidurus2;

Time in seconds for last procedure was 0.00

* * WARNING * * Value of LOTS defaults to 1


INPUT --- proc chmatrix occasions=132 groups=2 FPool;

INPUT --- glabel(1)=IBGE;

INPUT --- glabel(2)=JBB;

INPUT --- glabel(2)=JBB;

Length of command exceeded MAXCHR/2 characters.




What is this Length of command thing and how cani make the test work?

[prldb]

I have also just obtained this error message. I gather from a comment by Gary to a similar post that MARK has a limit to the number of characters it can read prior to the ';'. I am running the fully time-dependent CJS model with 18 capture occassions. No problem with running RELEASE without any group labels or with four group labels but got the error message with 16 group labels. There are no comment strings or other data in the row. Is it surprising that I received this error message with this input? Is there a work around other than reducing the number of group labels? Is there a similar limitation in U-Care?

[gwhite]

I think you have something else wrong, because you should not be receiving that message with only 18 occasions. Better carefully check to see that the semicolons are actually in place, and that the text is ASCII.

Gary

[prldb]

Wouldn't those kinds of syntax problems prevent MARK from running the CJS analysis (which ran okay) as well as RELEASE? I only got the error message when trying to run RELEASE.

[prldb]

In case this information should prove useful to someone else. I did get the error message due to a syntax problem with the inp file (apparently). Visual inspection of the inp file showed no problems with the file and the CJS analyses are performed normally by MARK. When I try to run RELEASE, however, the last few lines of the RELEASE output file are:

INPUT --- glabel(14)=VrMU;

INPUT --- glabel(15)=VrFR;

INPUT --- glabel(16)=VrFU;

INPUT --- glabel(16)=VrFU;

Length of command exceeded MAXCHR/2 characters.

The peculiar line is the repeated line refering to the 16'th group. I don't know why RELEASE repeats that line but presumably that's the glitch. A colleague imported my inp into Excel as space delimited (the column formats were brought in as text to keep the zeros), and then saved the imported file using the ‘Text (MS-Dos) (*.txt)’ option in save. I then changed the extension to inp and this file (without any editing) reproduced the same analysis results for CJS modelling and RELEASE produced no error message and ran normally. Bizzarely, when we tried to use U-CARE, U-CARE ran with the original version of the inp (which RELEASE complained about) but shut down when we tried to open the second version of the inp file (that RELEASE did not complain about).