Predicting the number of skilled emigrants in the Philippines: comparison of probabilistic forecasting methods

The Philippines is among the major exporters of skilled workers worldwide, and according to present data most of the professionals constitute a major group among Filipino migrant laborers. To understand this phenomenon of all the Filipino migrant laborers, this study aims to predict the number of skilled Filipino emigrants around the world using three probabilistic forecasting models such as ARIMA models, Econometric model with covariates, and Bayesian model. And the data being utilized in the study covers the period from 1981 to 2019 from the Open Data Philippines, specifically the Philippines Statistics Authority. In the analysis utilizing Bayesian ARIMA model it shows that it had the most accurate predictions with a lower root mean square error (RMSE) while incorporating prior distribution compared to other models, ARIMA model and Econometric model. Therefore, this study emphasizes the importance of considering appropriate probabilistic methods for accurate migration forecasting which can be significant especially to our policymakers, which is through this model they can address the brain drain issues and enhance workforce retention strategies.