generate_independent_covariates: Generate independent covariates - numerical or (ordinal)...

Description Usage Arguments Details

View source: R/features_independent.R

Description

Generate independent covariates - numerical or (ordinal) categorical

Usage

1
2
generate_independent_covariates(no_pts, no_covariates, type = "numerical",
  no_categories, ...)

Arguments

no_pts

integer; number of data points to be simulated.

no_covariates

integer; number of covariates.

type

One of 'numerical' or 'categorical'.

no_categories

integer; number of categories. Only applies when type = 'categorical'.

...

See details.

Details

If 'numerical' is chosen, one can specify the distribution through parameter 'distn'. The choices are "gaussian", "student-t" or "pareto". One should also provide the parameters when one specifies the distribution. For more details, see "?rnorm", "?rt", "?VGAM::rpareto". Note that for 'gaussian', standard normal is assumed if parameters are not given.

If 'categorical' is chosen, one can specify the distribution for the categories through the parameter 'prob'.


kcf-jackson/glmSimData documentation built on May 20, 2019, 8:15 a.m.