Package: SparseMSE 3.0.1

SparseMSE: 'Multiple Systems Estimation for Sparse Capture Data'

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) <https://www.tandfonline.com/doi/full/10.1080/01621459.2019.1708748> and in "Bootstrapping multiple systems estimates to account for model selection" Silverman, B. W., Chan, L, Vincent, K (2023)<https://doi.org/10.1007/s11222-023-10346-9>. 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.

Authors:Lax Chan [aut, cre], Bernard Silverman [aut], Kyle Vincent [aut]

SparseMSE_3.0.1.tar.gz
SparseMSE_3.0.1.zip(r-4.5)SparseMSE_3.0.1.zip(r-4.4)SparseMSE_3.0.1.zip(r-4.3)
SparseMSE_3.0.1.tgz(r-4.4-any)SparseMSE_3.0.1.tgz(r-4.3-any)
SparseMSE_3.0.1.tar.gz(r-4.5-noble)SparseMSE_3.0.1.tar.gz(r-4.4-noble)
SparseMSE_3.0.1.tgz(r-4.4-emscripten)SparseMSE_3.0.1.tgz(r-4.3-emscripten)
SparseMSE.pdf |SparseMSE.html
SparseMSE/json (API)

# Install 'SparseMSE' in R:
install.packages('SparseMSE', repos = c('https://laxchan.r-universe.dev', 'https://cloud.r-project.org'))

Peer review:

Bug tracker:https://github.com/laxchan/sparsemse/issues

Datasets:
  • Artificial_3 - Artificial data set to demonstrate possible instabilities
  • Korea - Korea data
  • Kosovo - Kosovo data
  • Ned - The Netherlands data
  • Ned_5 - Netherlands data five list version
  • NewOrl - New Orleans data
  • NewOrl_5 - New Orleans data five list version
  • UKdat - UK data
  • UKdat_5 - UK data five list version
  • Western - Victims related to sex trafficking in a U.S. Western site
  • hiermodels - Hierarchical models

On CRAN:

2.00 score 7 scripts 122 downloads 49 exports 2 dependencies

Last updated 11 months agofrom:627b8bbb2e. Checks:OK: 1 ERROR: 6. Indexed: yes.

TargetResultDate
Doc / VignettesOKNov 03 2024
R-4.5-winERRORNov 03 2024
R-4.5-linuxERRORNov 03 2024
R-4.4-winERRORNov 03 2024
R-4.4-macERRORNov 03 2024
R-4.3-winERRORNov 03 2024
R-4.3-macERRORNov 03 2024

Exports:ancestorsassemble_bicbcaconfvaluesBICandbootstrapsimbicktopahatcalBICrank_tiebreakbootstrap_msebootstrapcalboundary_capturesbuildmodelbuildmodelmatrixcheckallmodelscheckidentcheckident.1checkident.2checkthetasubsetchild_capturesconvert_from_hierarchyconvert_to_hierarchycount_triplesdecode_capturedescendantsdownhill_bootstrapcaldownhill_fitdownhill_funsdownhill_jackknifecalencode_captureestimatepopulationestimatepopulation.0find_bic_rank_matrixfind_neighbour_hierarchiesfind_unique_patternsfit_hier_modelgethiermodelsingest_datainvestigateAICjackknifecalktopBCamake_master_designmodelfitmodelorderntopBCaordercapturesparent_capturesremovenoninformativelistsstepwisefitsubsetmatsubsetsearchtidylists

Dependencies:lpSolveRcapture

Readme and manuals

Help Manual

Help pageTopics
Find the "ancestors" of a given capture historyancestors
Artificial data set to demonstrate possible instabilitiesArtificial_3
Models BICS, abundance and maxorder.assemble_bic
BCa confidence intervalsbcaconfvalues
Comparison of BIC approach and BCa approachBICandbootstrapsim
Acceleration factor calculationbicktopahatcal
Return a BIC rank matrix breaking ties with each rolesBICrank_tiebreak
A routine for naive userbootstrap_mse
Bootstrap abundance and bicbootstrapcal
Given a vector of captures, find those which are not in the vector but all of whose parents areboundary_captures
Build model for multiple systems estimationbuildmodel
Build the model matrix based on particular data, as required to check for identifiability and existence of the maximum likelihood estimatebuildmodelmatrix
Check all possible models for existence and identifiabilitycheckallmodels
Check a model for the existence and identifiability of the maximum likelihood estimatecheckident
The Fienberg-Rinaldo linear program check for the existence of the estimatescheckident.1
Carry out the Fienberg-Rinaldo procedure on an array of data vectors and a vector of modelscheckident.2
Check a subset of the parameter set thetacheckthetasubset
Find the "children" of a given capture historychild_captures
Find the vector of captures corresponding to a given hierarchical modelconvert_from_hierarchy
Find hierarchical representation of a vector of capturesconvert_to_hierarchy
Count number of triples of overlapping listscount_triples
Decode capture historydecode_capture
Find the "descendants" of a given capture historydescendants
Bootstrap downhilldownhill_bootstrapcal
Conduct downhill search among hierarchical models starting from the main effects only.downhill_fit
Downhill wrapperdownhill_funs
Jackknife downhilldownhill_jackknifecal
Encode capture historyencode_capture
Bootstrapping to evaluate confidence intervals using BCa methodsestimatepopulation
Estimate the total population including the dark figure. If the user wishes to find bootstrap confidence intervals then the routine 'estimatepopulation' should be used instead.estimatepopulation.0
Find BIC rank matrix up to a specified number of bic ranks for a given data setfind_bic_rank_matrix
Find all neighbouring hierarchical model to a given onefind_neighbour_hierarchies
Find unique patterns in matrix columnsfind_unique_patterns
Fit a hierarchical model taking account of possible sparsityfit_hier_model
Get a list of all hierarchical models for given number of lists and maximum ordergethiermodels
Hierarchical modelshiermodels
Preliminary processing of a data matrixingest_data
Plot of simulation studyinvestigateAIC
Jackknife abundance and Jackknife bicjackknifecal
Korea dataKorea
Kosovo dataKosovo
Find BCa confidence intervals using ktop idea.ktopBCa
Set up the inclusion matrix for all possible capture historiesmake_master_design
Fit a specified model to multiple systems estimation datamodelfit
Order of modelsmodelorder
The Netherlands dataNed
Netherlands data five list versionNed_5
New Orleans dataNewOrl
New Orleans data five list versionNewOrl_5
Find BCa confidence intervals for all possible ntopntopBCa
Order capture historiesordercaptures
Find the "parents" of a given capture historyparent_captures
Remove non-informative listremovenoninformativelists
Models BICS, abundance and maxorder.sortmodelsbic
Stepwise fit using Poisson p-values.stepwisefit
Subset matrixsubsetmat
Search subsets for a property which is inherited in a particular waysubsetsearch
Produce a data matrix with a unique row for each capture historytidylists
UK dataUKdat
UK data five list versionUKdat_5
Victims related to sex trafficking in a U.S. Western siteWestern