Backward covariate search
Usage
backwardSearch(
varsVec,
covarsVec,
catvarsVec = NULL,
fitorig,
fitupdated,
pVal = 0.01,
reFitCovars = FALSE,
outputDir,
restart = FALSE
)Arguments
- varsVec
character vector of variables that need to be added
- covarsVec
character vector of covariates that need to be added
- catvarsVec
character vector of categorical covariates that need to be added
- fitorig
the original 'fit' object before forward search
- fitupdated
the updated 'fit' object, if any, after the forward search
- pVal
p-value that should be used for selecting covariates in the forward search
- reFitCovars
if the covariates should be added before performing backward search - useful for directly performing backward search without forward search; default is FALSE
- outputDir
the name of the output directory that stores the covariate search result
- restart
a boolean that controls if the search should be restarted; default is FALSE