local.ml

A design matrix with cell counts (possibly fractional)
included in the column named "c".

pdat

A model specification such as "indep" or "indep.mono".

ml.meth

A character vector of predictors of the form "c1", "c2"
for main effects, or "c12" for an interaction.

predictors


A workhorse function for <code>apply.local.ml</code>.  For a fixed log-linear model,
get best fit using maximum local likelihood estimation.


Applies local log-linear capture-recapture models (LLLMs) for
closed populations, as described in the doctoral thesis of Zachary Kurtz.
The method is relevant when there are 3-5 capture occasions, with auxiliary
covariates available for all capture occasions.  As part of estimating the
number of missing population units, the method estimates the "rate of
missingness" as it varies over the covariate space.  In addition,
user-friendly functions are provided to recreate (approximately) the method
of Zwane and van der Heijden (2004), which applied the VGAM package
in a way that is closely related to LLLMs.

local.ml: Maximum likelihood estimation for fixed LLLMs

Description

Usage

Arguments

Details

References