pdc's Introduction

`pdc` Overview

Overview

The problem of model selection is inevitable in an increasingly large number of applications involving partial theoretical knowledge and vast amounts of information, like in medicine, biology or economics. The associated techniques are intended to determine which variables are "important" to "explain a phenomenon under investigation. The terms "important" and "explain" can have very different meanings according to the context and, in fact, model selection can be applied to any situation where one tries to balance variability with complexity.

This package implements the prediction divergence criterion model selection apporach proposed in Avella Medina et al., (2017).

Installation

This package can be installed as

# install.packages("devtools")
devtools::install_github("smac-group/pdc")

Features

TO DO

References

TO DO

pdc's People

Contributors

Watchers

pdc's Issues

Add possibility to use PDC when p > n

Extend pdc function by adding screening.

Add error messages

Add error messages for x_inter.

Ordering (adaptive/non-adaptive)

Marco could you please check if what I did is correct for adaptive/non-adaptive greedy? I'm guessing there is a smarter to do this? Could you please check? The code is in r/order.R and the part to check is below:

if (adaptive == TRUE){
      w <- as.vector(abs(solve(t(sX)%*%sX,t(sX)%*%y0)))
      norm.col <- as.vector(sqrt(t(rep(1,n))%*%sX^2))
      wX <- X%*%diag(w/norm.col)
   }else{
   wX <- X
}