Inflation forecasts using Bayesian Techniques
Uloženo v:
| Autoři: | , |
|---|---|
| Médium: | artículo original |
| Stav: | Versión publicada |
| Datum vydání: | 2015 |
| Popis: | The effective monetary policy using the inflation targeting scheme proposed by the Central Bank of Costa Rica is mostly based on the correct and timely inflation forecast in the short and medium term, in order to better design monetary policy actions. The purpose of this study is to develop a complementary tool to forecast inflation using a Bayesian approach. To that end, we propose using the Bayesian Model Averaging and Weighted-Average Least Squares methodologies. Such projection models allow expanding and complementing the analysis currently conducted by the Central Bank of Costa Rica using the Quarterly Macroeconomic Projection Model (MQPM). As a result, we show that for monthly data and forecasting 1 to 12 months in advance, it is possible to have projections using the Bayesian process with greater predictive performance than with the autoregressive model. |
| Země: | Portal de Revistas UNA |
| Instituce: | Universidad Nacional de Costa Rica |
| Repositorio: | Portal de Revistas UNA |
| Jazyk: | Español |
| OAI Identifier: | oai:ojs.www.una.ac.cr:article/7287 |
| On-line přístup: | https://www.revistas.una.ac.cr/index.php/economia/article/view/7287 |
| Klíčové slovo: | productive development entrepreneurs methodology of extensiveness social vulnerability Modelos de series temporales números índice y agregación predicción y simulación Análisis bayesiano. |