Estimability problems with MState and missing data

[nachoman]

Hi
I am running a HFHetRDMS analysis in RMark using 740ish capture histories with 18 primary occassions*5 secondary occassions each. I have an issue because I have not data for some of the primary occassions. On Year1, I captured in May, June, July, Aug, Oct and same months on Year2 (e.g. I missed September both years). Following Cooch's advice through this forum, I have coded missing captures as dots (".") to preserve all intervals between captures equal. Estimates of monthly transitions and survival rates are calculated fine for May-June, June-July and July-Aug both years, but (unsurprisingly) NOT for Aug-Sept or Sept-Oct or any of the winter Oct-Nov Year1, etc...

I have tried several tricks to circumvent this. First, I have fixed p=0 for missing primary occasions, which actually produces more acurate estimates of May-June, June-July and July-Aug estimates, but no effect on the problematic intervals. I have also tried defining new bins merging some betas so that no beta estimates have missing primary occasions at any end of its corresponding interval, e.g. simultaneously or separately defining bin1=(Aug-Sept year1 + Sept-Oct year1), bin2=(Aug-Sept year2 + Sept-Oct year2), bin3=(all winter Year 1 + spring Year2 monthly intervals). I intuitively see that for bin3 it's difficult to obtain accurate estimates because there is plenty of missing data, but I don't see why it is so hard to obtain estimates of the other two bins constraining parameters in that way.

Ideally I would like at least to obtain reasonable estimates for all monthly intervals between May and October, and not throw to the bin the October primary capture data...I have thought about trying the old trick of recoding the capture history without missing data and with unequal time intervals, which would allow to calculate S, but obviously I would'nt be able to get meaningful Psis in that way.

I would appreciate any advice from more experienced users. Cheers,

N