This repository contains the results, and the scripts that generated those results, for a study comparing the power of dynamic and static risk factors to predict recidivism in Finland. We use elastic net logistic regression and random forest as supervised learning methods and allow for the possibility that different sets of predictors work better with either algorithm.
A preprint is available at https://psyarxiv.com/v5uq3/ . The current manuscript is a revised version that is currently being rewritten according to reviewer suggestions.
Files are organised in an R package structure (even though currently it is not designed to be installed in R). The following specification might help:
The manuscript specifically mentions the following to be found on these pages
(Updated 2019-03-13)
Add the following code to your website.
For more information on customizing the embed code, read Embedding Snippets.