modslavmse-package: 'modslavmse'

In bernardsilverman/modslavmse: Multiple Systems Estimates for estimating the prevalence of Modern Slavery

Description

modslavmse is a package for Multiple Systems Estimation as applied particularly in the context of Modern Slavery

Details

This package serves three purposes. Firstly, it implements the method used in the original work by Bales, Hesketh and Silverman to approach the question of the prevalence of Modern Slavery in the UK by estimating the 'dark figure' underlying data collected in the 2013 National Referral Mechanism (NRM) by the National Crime Agency (NCA). That work used the package Rcapture with a particular approach to the choice of two-factor interactions to fit in the model. This method is implemented in the routine MSEfit. The data set UKdat gives the data used in the original paper. The research also involved testing the robustness of the results by omitting some of the lists and/or combining some into single lists. The routines omitlists and mergelists allow these operations to be done.

The other purpose of this package is to implement development versions of current research directions. A current focus is on Monte Carlo Markov Chain approaches which allow some sort of model averaging, rather than the focus on a particular model implicit in MSEfit. This makes use of the package MCMCpack.

The third, more specific, purpose is to allow full reproducibility of the work presented in Silverman (2018). This is done through the scripts given in the Examples section below.

Author(s)

Maintainer: Bernard W. Silverman [e-mail]

References

K. B. Bales, O. Hesketh, and B. W. Silverman (2015). Modern Slavery in the UK: How many victims? Significance 12 (3), 16-21.

B. W. Silverman (2018). Model fitting in Multiple Systems Analysis for the quantification of Modern Slavery: Classical and Bayesian approaches.

Examples

data(UKdat, UKdat_5, UKdat_4, Ned, Ned_5, NewOrl, NewOrl_5, Kosovo) # the datasets used in the paper
make_AIC_stepwise_table1()  # Table 5
make_AIC_stepwise_table2()  # Table 6
make_allmodels_plots_script() # Figures 1, 2, 3 and 4
make_MCMCfit_tables_script() # Tables 7, 8, 9, 10, 11 and 12
make_MCMCeffects_table_script() #   Table 13
make_madyor_table_script()     #  Table 14 and Figures 5 and 6, as well as some numbers in the text
make_LCMCR_table_script()      # Table 15

bernardsilverman/modslavmse documentation built on Aug. 10, 2019, 2:39 a.m.