Comparative Study of Time Domain Models in the Analysis of Aggregated Crime Rate in Nigeria

Adeboye, Nurain Olawale and Omosanya, K. O and Agunbiade, Dawud Adebayo (2018) Comparative Study of Time Domain Models in the Analysis of Aggregated Crime Rate in Nigeria. In: 2nd NSS International Conference, 2018, University of Calabar, Nigeria.

[img] Text
P448-452_PSSN-CP-2018.pdf

Download (512kB)

Abstract

The high rate of crime in Nigeria is a serious concern for the government and individuals because of its economic implication, serious damages and likely loss of lives and properties. Nigeria is one of the countries with a high rate of crime because of unemployment, economic hardship and crime friendly environment. It has, therefore, become imperative to ascertain the appropriate time-domain modeling for the crime rate in the country. As a result of this, ARIMA and SARIMA models were examined on the aggregated crime committed in Nigeria between the years 2005 and 2015. It was discovered that the data was not stationary and stationarity was achieved through 2nd order differencing and 2nd order seasonal differencing for ARIMA and SARIMA models respectively. Augmented Dickey- Fuller test was then performed on the differenced series in order to check for its acceptability. The AIC, BIC, AICc, AME and RMSE were used to select the models that really provided the best fit for the series and it revealed that both ARIMA (5,2,4) and SARIMA (4,2,3)(1,2,1)12 are the best. The chosen SARIMA model, however, outclassed the ARIMA model due to its overall lower values of model selection criteria used and its better forecasting accuracy. The chosen model was then used to forecast the crime rate in Nigeria for five years and the results revealed an expectation of a downward trend in crime rate in Nigeria.

Item Type: Conference or Workshop Item (Paper)
Subjects: Q Science > Q Science (General)
Divisions: Faculty of Engineering, Science and Mathematics > School of Mathematics
Depositing User: Mr Taiwo Egbeyemi
Date Deposited: 13 Jul 2020 11:39
Last Modified: 13 Jul 2020 11:39
URI: http://eprints.federalpolyilaro.edu.ng/id/eprint/1009

Actions (login required)

View Item View Item