search function

Stepwise Selection Search

To search itemset that give maximum value of the criterion

stepwise_search(
  X,
  criterion = c("ipoqll", "ipoqlldif"),
  incl_set = c(),
  groups_map = c(),
  cores = NULL,
  isContinued = FALSE,
  prevData = c(),
  fileOutput = FALSE,
  tempFile = "temp_stepSearch.RData",
  isConvert = FALSE,
  setting_par_iq = c(),
  setting_par_oq = c(),
  method = c("fast", "novel"),
  isTraced = FALSE
)

backward_search(
  X,
  criterion = c("ipoqll", "ipoqlldif"),
  incl_set = c(),
  groups_map = c(),
  cores = NULL,
  isContinued = FALSE,
  prevData = c(),
  isConvert = FALSE,
  setting_par_iq = c(),
  fileOutput = FALSE,
  setting_par_oq = c(),
  method = c("fast", "novel"),
  tempFile = "temp_backSearch.RData",
  isTraced = FALSE
)

## S3 method for class 'search'
summary(object, ...)

## S3 method for class 'search'
print(x, ...)

plot_search(obj, remOrdered = TRUE, locateMax = TRUE, ...)

Arguments

• X: A matrix or data.frame of the observed responses (ordinal or binary response).
• criterion: The criterion that should be used. The default is ipoqll.
• incl_set: A vector of initial items in the included set to start the search. The default is to start with full items.
• groups_map: A matrix or vector to map the subject to the DIFs groups.
• cores: An integer value of number of cores should be used for computation. The default is 2.
• isContinued: A logical value whether this search is continuing another unfinished search.
• prevData: The filename of the temporary .RData file of the unfinished search.
• fileOutput: The filename if it is wished to save the output results in file (.RData and .csv) and FALSE if not.
• tempFile: The filename of the temporary file to track the search progress. The default is "temp_stepSearch.RData" which also automatically produces "temp_stepSearch.csv".
• isConvert: A logical value whether it is wanted to recompute the score of the search results using IPOQ-LL-DIF criterion.
• setting_par_iq: a list of the optimization control setting parameters for the included set. See setting parameter in autoRaschOptions().
• setting_par_oq: a list of the optimization control setting parameters for the included set. See setting parameter in autoRaschOptions().
• method: The implementation option of log likelihood function. fast using a c++ implementation and novel using an R implementation.
• isTraced: A logical value whether the progress need to be tracked or not.
• object: The object of class 'search'.
• ...: Further arguments to be passed.
• x: The object of class 'search'.
• obj: An object of class "search".
• remOrdered: A logical statement whether show the order of the items removal or not.
• locateMax: A logical statement whether the location of the maximum score is needed to be marked or not.

Returns

Matrix of the highest scores (IQ-LL, OQ-LL, and IPOQ-LL) for every number of items in the included set in the set along with the corresponding itemset.

Details

To search the itemset that give the maximum score.

Examples

## Not run:

search_res <- backward_search(shortDIF,criterion = "ipoqll", incl_set = c(1:4), cores = 2)
plot_search(search_res, type="l")
## End(Not run)