Time Series Regression Models and Trends in Drug, Substance Abuse and Index Crimes

Abstract

It is vital for the Philippines to understand the time series interrelationship of index crimes and drug/substance abuse amidst its mandate on the “War on Drugs”. The purpose of this paper is to investigate the best fitting forecasting models of the time series trends of index crimes and drug/substance abuse in the Philippines. The study employed three approaches; (1) delineating time series regression models of index crimes with drug/substance abuse as explanatory variables, (2) describing pure time series analysis by ways of autoregressive integrated moving average (ARIMA) and exponential smoothing, and (3) forecasting the time series trends from 2018 to 2022. These were carried out using IBM SPSS professional. The findings of this study demonstrate an association between drug/substance abuse and index crimes. The regression analyses concluded several associations. For example, variations of the time series data in murder can be predicted by reported new and re-admission cases to various rehabilitation centers and reported cases of marijuana abusers. Rape cases are associated with the admission to rehabilitation centers and reported Shabu abusers. Lastly, carnapping is explained by the number of cases of Shabu abusers. Although some research on the relationship between crime and drug abuse is yet to be done, this novice idea on model delineation in its time series trends will be of immense help among policymakers.