I'm running a POPAN model for analysis of fish mark-recap. The model is specified as follows:
Phi.Weight = list(formula = ~ Weight)
pent.Country = list(formula = ~ Country)
p.t.x.Country = list(formula=~ Country*time)
N.Country = list(formula = ~ Country)
where Weight is individual fish weight at release, and Country is country of release. The model runs great, but when I try to use popan.derived(), I get the error "Error in covariate.predictions(model, data.frame(index = index), drop = drop) : Computations for mlogit parameters with covariate values cannot be specified with index column in data; use separate indices argument".
If I run the model with intercept-only phi, pent, and N, the popan.derived() runs perfectly, but it's a worse model... I haven't been able to find similar issues on the web, or the error message within popan.derived() itself, so must be an error within a nested function, so I'm lost. Any help would be appreciated.