TO GET THE COMPLETE JOURNAL/THESIS FOR TOPIC BELOW,
CALL: 08168759420, 08068231953
WHATSAPP: 08137701720
THE IMPACT OF COVID-19 ON UNEMPLOYMENT RATE: AN INTELLIGENT BASED UNEMPLOYMENT RATE PREDICTION IN SELECTED COUNTRIES OF AFRICA
Abstract
Unemployment remains a major cause for both developed and developing nations, due to which they lose their financial and economic impact as a whole. Unemployment rate prediction achieved researcher attention from a fast few years. The intention of doing our research is to examine the impact of the coronavirus on the unemployment rate. Accurately predicting the unemployment rate is a stimulating job for policymakers, which plays an imperative role in a country’s financial and financial development planning. Classical time series models such as ARIMA models and advanced non-linear time series methods be previously hired for unemployment rate prediction. It is known to us that mostly these data sets are non-linear as well as non-stationary. Consequently, a random error can be produced by a distinct time series prediction model. Our research considers hybrid prediction approaches supported by linear and non-linear models to preserve forecast the unemployment rates much precisely. These hybrid approaches of the unemployment rate can advance their estimates by reproducing the unemployment ratio irregularity. These models’ appliance is exposed to six unemployment rate statistics sets from Africa’s selected countries, specifically France, Spain, Belgium, Turkey, Italy and Germany. Among these hybrid models, the hybrid ARIMA-ARNN forecasting model performed well for France, Belgium, Turkey and Germany, whereas hybrid ARIMA-SVM performed outclass for Spain and Italy. Furthermore, these models are used for the best future prediction. Results show that the unemployment rate will be higher in the coming years, which is the consequence of the coronavirus, and it will take at least 5 years to overcome the impact of COVID-19 in these countries.
K E Y W O R D S
Abbreviations: ACF, autocorrelation function; AIC, Akaike information criterion; ANN, artificial neural networks; ARIMA, autoregressive integrated moving average; ARNN, autoregressive neural network; BIC, Bayesian information criterion; MAE, mean absolute error; MAPE, mean absolute percent error; PACF, partial autocorrelation function; RMSE, root mean square error; SVM, support vector machine.