Modeling Unemployment Rates in Nigeria using Time Series Approach

Unemployment is a social-economic situation that has resulted into poor standard of living, increases in crime rate and insecurity in the country. It even possesses greater threats to its development and peaceful co-existence. It is in light of proffering solutions to the problems of unemployment that this study is germane. We examined the trend, model and forecast for the future rate of unemployment in Nigeria. The statistical methodology adopted in this research work is mainly univariate time series analysis using autoregressive integrated moving averages (ARIMA).The time plot showed that rate of unemployment in Nigeria is not stable over the years. ADF test was performed which confirms the non-stationarity of the process (P > 0.05). The data was transformed through differencing. KPSS test was used to confirm the stationarity of the process (P > 0.05).Various ARIMA (p,d,q) were examined such as ARIMA (1,1,1), ARIMA (2,1,1), ARIMA (2,1,2), ARIMA (2,2,1) and ARIMA (2,2,2) with AICs and Log-likelihood (107.52, −50.76), (109.22, −50.61), (111.08, −50.54), (110.08, −51.04) and (111.36, −50.68) respectively. ARIMA (1,1,1) was selected due to its least AIC and highest log likelihood. Both Shapiro-Wilk test and Box test performed confirm the fitness of the model (P > 0.05) for the process. Forecast for 5 years was then made for the process. In conclusion, the model obtained in this research can be used for making inference, monitor and control unemployment rates in Nigeria. Adequate solutions such as industrialization, diversification in the economy, investment in agriculture, modification in education curriculum among others have been identified in this research work. The analysis is completely done using R package