Forecasting of Annual Crime Rate in India: A case Study

Crime forecasting has been made possible by criminological theories and developments in computational techniques/data analytics further improved the technologies of forecasting. It helps the various departments of the police to make a decision and strategy to prevent the crime. This paper focuses on forecasting the annual crime rate in India using the Time Series Models such as Auto-Regressive Integrated Moving Average (ARIMA) and Exponential Smoothing. Source of data is from the National Crime Record Bureau of India. As a part of modeling, data is divided into training data for the years 1953 to 2008 and test data for the years 2009 to 2013. By examining the model, it's clear that the forecast values are within the 95% confidence interval of the test data and accuracy measurements are also significant. Hence the time series model suitable for crime forecasting.