stablespec: Stable Specification Search in Structural Equation Models
Version 0.3.0

An exploratory and heuristic approach for specification search in Structural Equation Modeling. The basic idea is to subsample the original data and then search for optimal models on each subset. Optimality is defined through two objectives: model fit and parsimony. As these objectives are conflicting, we apply a multi-objective optimization methods, specifically NSGA-II, to obtain optimal models for the whole range of model complexities. From these optimal models, we consider only the relevant model specifications (structures), i.e., those that are both stable (occur frequently) and parsimonious and use those to infer a causal model.

AuthorRidho Rahmadi [aut, cre], Perry Groot [aut, ths], Tom Heskes [aut, ths], Christoph Stich [ctb]
Date of publication2017-04-05 03:27:52 UTC
MaintainerRidho Rahmadi <r.rahmadi@cs.ru.nl>
LicenseMIT + file LICENSE
Version0.3.0
URL https://github.com/rahmarid/stablespec
Package repositoryView on CRAN
InstallationInstall the latest version of this package by entering the following in R:
install.packages("stablespec")

Popular man pages

crossdata6V: Artificial cross-sectional data.
dataReshape: Reshape longitudinal data
getModelFitness: Scoring the given SEM models.
longiData4V3T: Artificial longitudinal data.
modelPop: Random SEM models.
plotStability: Plot of edge and causal path stability.
repairCyclicModel: Repairing a SEM model that is cyclic.
See all...

All man pages Function index File listing

Man pages

crossdata6V: Artificial cross-sectional data.
dataReshape: Reshape longitudinal data
getModelFitness: Scoring the given SEM models.
longiData4V3T: Artificial longitudinal data.
modelPop: Random SEM models.
plotStability: Plot of edge and causal path stability.
repairCyclicModel: Repairing a SEM model that is cyclic.
stableSpec: Stable specifications of constrained structural equation...

Functions

causalCounter Source code
cons4Stab Source code
convertCons Source code
convertFront Source code
crossOver Source code
crossdata6V Man page
crowdingDistance Source code
cycleRepair Source code
dag2CpdagCons Source code
dataReshape Man page Source code
edgeCounter Source code
fastNonDominatedSort Source code
gatherFitness Source code
genPopulation Source code
getDataCross Source code
getDataLongi Source code
getFitness Source code
getMinBic Source code
getModelFitness Man page Source code
groupFrontComp Source code
initialPopulation Source code
longiData4V3T Man page
longiMatrixFill Source code
makeEdgeDf Source code
modelPop Man page Source code
mutation Source code
optimalModels Source code
orderEdge Source code
plotStability Man page Source code
relevantStructure Source code
repairCyclicModel Man page Source code
sameValue Source code
sortBasedOnDist Source code
stableSpec Man page Source code
stringToMatrix1 Source code
structureStab Source code
toDirect Source code
toLabelEdge Source code
toSkeleton Source code
topoOrder Source code
writeModel Source code

Files

tests
tests/testthat.R
tests/testthat
tests/testthat/test-modelFitness.R
tests/testthat/test-modelPop.R
tests/testthat/test-stablespec.R
tests/testthat/test-repairCycle.R
NAMESPACE
data
data/crossdata6V.rda
data/longiData4V3T.rda
R
R/longiData4V3T.R
R/getModelFitness.R
R/dag2CpdagCons.R
R/crowdingDistance.R
R/stableSpec.R
R/plotStab.R
R/toLabelEdge.R
R/crossdataV6.R
R/mutation.R
R/genPopulation.R
R/optimalModels.R
R/crossOver.R
R/writeModel.R
R/NSGAStuff.R
R/modelFitness.R
R/dataReshape.R
R/stringToMatrix.R
R/initialPopulation.R
R/edgeOperation.R
R/consOperation.R
R/groupFrontComp.R
R/makeEdgeDf.R
R/relevantStructure.R
R/repairCyclicModel.R
R/getData.R
R/modelPop.R
R/orderEdge.R
R/topoOrder.R
R/structureStab.R
MD5
DESCRIPTION
man
man/plotStability.Rd
man/stableSpec.Rd
man/repairCyclicModel.Rd
man/longiData4V3T.Rd
man/getModelFitness.Rd
man/crossdata6V.Rd
man/dataReshape.Rd
man/modelPop.Rd
LICENSE
stablespec documentation built on May 19, 2017, 11:37 p.m.

Questions? Problems? Suggestions? Tweet to @rdrrHQ or email at ian@mutexlabs.com.

Please suggest features or report bugs in the GitHub issue tracker.

All documentation is copyright its authors; we didn't write any of that.