VAR, ARIMAX and ARIMA models for nowcasting unemployment rate in Ghana using Google trends

Abstract The analysis of the high volume data spawned by web search engines on a daily basis allows scholars to scrutinize relation between user’s preferences and impending facts. This study can be used in variety economics contexts. purpose this is determine whether it possible anticipate unemployment rate examining behavior. method uses cross-correlation technique combine from Google Trends with World Bank's rate. Autoregressive Integrated Moving Average (ARIMA), eXogenous variables (ARIMAX) Vector Autoregression (VAR) models for prediction are fit using analyzed data. were assessed various evaluation metrics mean absolute error (MAE), root square (RMSE), percentage (MAPE), median (MedAE), maximum (ME). average outcome proved significant performance models. ARIMA (MSE = 0.26, RMSE 0.38, MAE 0.30, MAPE 7.07, MedAE 0.25, ME 0.77), ARIMAX 0.22, 0.29, 6.94, 0.75), VAR 0.09, 0.20, 4.65, 0.42) achieved margins. demonstrates that estimators improved reduction across board when compared model without them.