Return.key.IPD.summaries: Computation of input IPD summaries from the original IPD

In bonorico/gcipdr: Gaussian Copula (based) Individual Person Data (IPD) Reconstruction

Description

This function computes marginal moments up to fourth degree, all pairwise correlation of an available IPD and Johnson system parameters for all IPD non-binary marginals. The IPD is supposed to be in a specific format (see 'Return.IPD.design.matrix()' and Details of 'DataRebuild()').

Usage

Return.key.IPD.summaries(md, corrtype = c("rank.corr", "moment.corr",
  "normal.corr"), only.fixed.stats = FALSE, na.erase = TRUE)

Details

Ideally the the IPD holder uses this function output as a template for the IPD summaries format to be disclosed and needed in 'DataRebuild()' by a IPD summary holder. If IPD has missing data, the module can compute IPD summaries as if missing records were imputed. This must be decided before via argument fill.missing in 'Return.IPD.design.matrix()'.

See Also

[Return.IPD.design.matrix()] for currently requested IPD format, [DataRebuild()]

bonorico/gcipdr documentation built on May 2, 2021, 8:12 p.m.