Prediction modelling for population conviction data

Abstract

This monograph is concerned with different uses of prediction modelling on (Dutch) Judicial registration data. It covers the determining the usefulness, preservability of a prediction instrument used in the judicial context, determining the prediction model that optimally predicts different types of recidivism using classical and modern models differing in flexibility, and the estimation of the effect of an intervention in a missing values and observational data context. CONTENT: 1. Introduction 2. StatRec - performance, validation and preservability of a static risk prediction instrument 3. Comparing predictive performance of statistical, machine learning and data mining predictive models on reconviction data 4. Searching for improvements in predictive performance by applying machine learning in binary outcome and censored recidivism data 5. Using combinations of multiple inputation, propensity score matching and difference-in-differences for estimating the effectiveness of a prolongued incarceration and rehabilitation measure for high-frequent offenders