Nowcasting domestic liquidity in the Philippines using machine learning algorithms

This study utilizes a number of algorithms used in machine learning to nowcast domestic liquidity growth the Philippines. It employs regularization (i.e., Ridge Regression, Least Absolute Shrinkage and Selection Operator (LASSO), Elastic Net (ENET)) tree-based Random Forest, Gradient Boosted Trees) methods order support BSP’s current suite macroeconomic models forecast analyze liquidity. Hence, this evaluates accuracy time series (e.g., Autoregressive, Dynamic Factor), regularization, through an expanding window. The results indicate that LASSO, ENET, Trees provide better estimates than traditional models, with month-ahead nowcasts yielding lower Root Mean Square Error (RMSE) (MAE). Furthermore, facilitate identification indicators are significant specify parsimonious nowcasting models.