Individual Covariates -- Missing Values

Individual covariates are entered into Program MARK via the Encounter Histories File. These values follow the number of animals encountered for the encounter history, as described in the Encounter Histories Format. Individual covariates are used to incorporate information specific to the individual into the parameter estimate. Individual covariates are not allowed to have missing values, because the parameter value for the animal with the missing individual covariate value would also be missing, and hence the animal could not contribute to the likelihood. Several options for handling individual covariates are available.

Probably the best option is to code missing individual covariate values with the mean of the variable for the population measured. Replacing the missing value with the average means that the mean of the observed values will not change, although the variance will be slightly smaller because all missing values will be exactly equal to the mean and hence not variable.

If you have lots of missing values, another option is to code the animals into 2 groups, where all the missing values are in one group. Then, you can use both groups to estimate a common parameter, and only apply the individual covariate to one group. This approach can be tricky, so think through what you are doing before you try this approach.