Simulating mock data

[szymekdr]

TO DO:

• [x] Determine order of simulating predictors based on the dependencies
• [x] Different type of predictors = different underlying distributions
• [x] Checks: predictor number agrees with order of simulation + sizes of X design matrices
• [x] Tree and species: arbitrary number of species m < n = final data size (species replicated at random) (now m=75, n=200)
• [x] generate final response based on the assumed linear model

[szymekdr]

User supplies:

#' @param n_species Size of the tree
#' @param n_cases Size of the final dataset
#' @param predictor_types A character vector of types of predictors
#'     c("gaussian", "binary", "poisson", "categoricalN"); for "categoricalN" - N defines
#'     the number of distinct categories to generate
#' @param response_type The distribution family for response variable
#' @param sim_matrix A matrix with ncol=nrow=length(predictor_types) filled with 0 and 1
#'     1 in row x and column y means predictor x depends on predictor y
#' @param var_names Optional character vector of variable names (if NULL
#'     generic names are used)
#' TBC...

[szymekdr]

• [ ] Package into clean functions and document