Forecasting the overall number of felony offenders in the State of Kuwait using time series models

Abstract

Time Series Analysis Techniques are fundamental in developing effective strategies for security planning, particularly in the field of crime prevention and control, as these methods significantly support security decision-making processes. This study aims to explore the application of time series models, with a focus on the ARIMA model, to forecast the total number of felony offenders in the State of Kuwait. The researcher relies on annual data published by the Research and Studies Center of Kuwait Ministry of Interior, covering the period from 1992 to 2022. The performance of various ARIMA models was evaluated using R-squared, Bayesian Information Criterion – Normalized (NBIC), and Root Mean Squared Error (RMSE) criteria. The study concluded that the optimal model is ARIMA (1,0,0). Based on this model, forecasts were made for the total number of felony offenders for the period from 2023 to 2030.