Function for computing estimates under a Weighted Rank Squared Error Loss function, with the option of parallelising intensive computations. Please see the accompanying notebook https://nbviewer.ipython.org/github/andrewzm/WRSEL/blob/master/ipython/WRSEL_Example_1.ipynb and (after having read the notebook!) the associated RShine application at https://andrewzm.shinyapps.io/WRSEL_example/ for an explanation of WRSEL and an interactive toy study.
This package is still under development and is not available on CRAN. To install you will need to have devtools installed and loaded. Then type in install_github('andrewzm/WRSEL'). Please do not hesitate to contact me if you have any queries.
Package: WRSEL
Type: Package
Title: Weighted Rank Squared Error Loss
Version: 1.0
Date: 2014-09-18
Author: Andrew Zammit Mangion
Maintainer: Andrew Zammit Mangion <azm [@] uow.edu.au>
Description: Find the optimal estimate under a weighted rank squared error loss (WRSEL) function. For details see Wright, Deanne L., Hal S. Stern, and Noel Cressie. "Loss functions for estimation of extrema with an application to disease mapping." Canadian Journal of Statistics 31.3 (2003): 251-266.
Suggests: parallel
Depends: R (>= 2.14)
Imports: Matrix
License: GPL (>= 2)
NeedsCompilation: no