Crime Prediction Based on Statistical Models

Abstract

Increasing rate of different crimes are always matter of concern in a peaceful society. Advancements in the field of information technology, publicly available information and services, somehow help criminals to achieve their misdeeds and involve them in much serious crimes than earlier. As a result, crime rate is increasing with a very high rate in developed and under-developed nations. Based on the previous year (s) crime details in Indian states, we present statistical models viz. Weighted Moving Average, Functional Coefficient Regression and Arithmetic-Geometric Progression based prediction of the crime in coming years. The crime details of Indian states between the years 2001 and 2013 have been used for this purpose. Crime details between 2001 and 2011 have been used for predicting the crimes for the years 2012 and 2013. These predicted values have been compared with actual crime details in the years 2012 and 2013. Difference between actual records and our predicted values for both years gives the accuracy of the proposed approaches between the range 85 % and 90%.