Modeling Nigeria’s Unemployment Rates: Box-Jenkins Approach

Abstract

One of the key determinants of economic growth and stability of a nation is the macroeconomic policy of a government. In Nigeria, despite the macroeconomic reforms and prudent policies pursued by the government, such reforms and policies have not translated to increase in employment opportunities and poverty reduction. The aim of this paper is to find the most suitable time series model of annual unemployment rates in Nigeria using Box-Jenkins methodology and to examine the precision of forecast on this model. ARIMA (2, 2, 1), ARIMA (2, 2, 2), ARIMA (1, 2, 1) and ARIMA (1, 2, 2) were fitted to the unemployment data. The Akaike Information Criterion and the Bayesian information Criterion suggested ARIMA(1,2,1) as the best model, however, evaluating a five year forecast accuracy using different measures of accuracy suggested that ARIMA(2,2,2)  minimizes the forecast error for the unemployment data. The projection made from 2020 to 2024 shows that unemployment rate will constantly increase in Nigeria if government at all levels do not implement economic policies that will translate to reduction in the unemployment rate. This projection does not take into cognizance any impact or effect of COVID-19 pandemic.