Forecasting Inflation in a data-rich environment: the benefits of machine learning methods
-
Required packages:
- HDeconometrics (from my github)
- h20
- glmnet
- xgboost
- randomForest
- lbvar (from my github)
- boot
-
This repository contains the codes used to run the rolling windows in the paper "Forecasting Inflation in a data-rich environment: the benefits of machine learning methods" by Medeiros, Vasconcelos, Veiga and Zilberman.
-
Codes are divided in one folder for each subsample used in the paper.
-
Each folder contains one folder with functions and one folder with codes to call and run the functions. PATHS MUST BE ADJUSTED.
-
If the paths are correct, running the files in the RUN folders will start the rolling windows.
-
Data are already treated as descriped in the paper.
Medeiros, M. C., Vasconcelos, G., Veiga, A., & Zilberman, E. (2018). Forecasting Inflation in a data-rich environment: the benefits of machine learning methods.