'Multiple Systems Estimation for Sparse Capture Data'
BCa confidence intervals
Comparison of BIC approach and BCa approach
Build model for multiple systems estimation
Build the model matrix based on particular data, as required to check ...
Check all possible models for existence and identifiability
Check a model for the existence and identifiability of the maximum lik...
Check a subset of the parameter set theta
Count number of triples of overlapping lists
Estimate the total population including the dark figure. If the user w...
Bootstrapping to evaluate confidence intervals using BCa methods
Plot of simulation study
Fit a specified model to multiple systems estimation data
Order capture histories
Remove non-informative list
Stepwise fit using Poisson p-values.
Search subsets for a property which is inherited in a particular way
Produce a data matrix with a unique row for each capture history
Implements the routines and algorithms developed and analysed in "Multiple Systems Estimation for Sparse Capture Data: Inferential Challenges when there are Non-Overlapping Lists" Chan, L, Silverman, B. W., Vincent, K (2019) <arXiv:1902.05156>. This package explicitly handles situations where there are pairs of lists which have no observed individuals in common. It deals correctly with parameters whose estimated values can be considered as being negative infinity. It also addresses other possible issues of non-existence and non-identifiability of maximum likelihood estimates.