by Carlos Espino, Xavier Gonzalez, Diego Llarrull, Woojin Kim
Understanding the factors behind criminal behaviour is one of the most crucial tasks for preventing and controlling future incidences of crime. In the included report, we explore the potential factors affecting crime rates based on the demographics and econometrics data gathered from 197 counties in North Carolina from 1981 to 1987. Using various statistical methods and modeling techniques, we analyze and identify the most important factors and metrics tied to crime rates. We also present a predictive model capable of estimating the crime rate with around 20% approximation error using the selected parameters.
The dataset was obtained from the Ecdat package for R.
Prepared for the STAT W4702: Statistical Inference and Modeling class, Fall 2015