IDEAS home Printed from https://ideas.repec.org/p/apk/doctra/2002.html
   My bibliography  Save this paper

Forecasting Costa Rican Inflation with Machine Learning Methods

Author

Listed:
  • Adolfo Rodríguez-Vargas

    (Department of Economic Research, Central Bank of Costa Rica)

Abstract

We present a first assessment of the predictive ability of machine learning methods for inflation forecasting in Costa Rica. We compute forecasts using two variants of K-Nearest Neighbours, random forests, extreme gradient boosting and a long short-term memory (LSTM) network. We evaluate their properties according to criteria from the optimal forecast literature, and we compare their performance with that of an average of univariate inflation forecasts currently used by the Central Bank of Costa Rica. We find that the best-performing forecasts are those of LSTM, univariate KNN and in lesser extent random forests. Furthermore, a combination performs better than the individual forecasts included in it and the average of the univariate forecasts. This combination is unbiased, its forecast errors show appropriate properties, and it improves the forecast accuracy at all horizons, both for the level of inflation and for the direction of its changes. ***Resumen: Se presenta una primera evaluación de la capacidad de métodos de aprendizaje automático para predecir la inflación en Costa Rica. Se calculan pronósticos mediante dos variantes de K-Nearest Neighbours (KNN), bosques aleatorios, extreme gradient boosting y un modelo de tipo long short-term memory (LSTM). Sus propiedades se evalúan de acuerdo con criterios sugeridos en la literatura sobre pronósticos óptimos, se compara su desempeño con el del promedio de los pronósticos univariados actualmente en uso en el Banco Central de Costa Rica. Los resultados son promisorios. Se encontró que los pronósticos con el mejor desempeño son los resultantes de aplicar LSTM, KNN univariado y en menor medida bosques aleatorios. Además, una combinación de los pronósticos obtenidos mediante estos métodos mejora el desempeño con respecto a los pronósticos individuales a todos los horizontes, y también supera en desempeño al promedio de los pronósticos univariados. La combinación resulta insesgada, sus errores de pronóstico no muestran patrones de correlación indeseables, y mejora la capacidad de pronóstico a todos los horizontes, tanto para el nivel de la inflación como para la dirección de sus cambios.

Suggested Citation

  • Adolfo Rodríguez-Vargas, 2020. "Forecasting Costa Rican Inflation with Machine Learning Methods," Documentos de Trabajo 2002, Banco Central de Costa Rica.
  • Handle: RePEc:apk:doctra:2002
    as

    Download full text from publisher

    File URL: https://repositorioinvestigaciones.bccr.fi.cr/handle/20.500.12506/335
    Download Restriction: no
    ---><---

    More about this item

    Keywords

    ;
    ;
    ;
    ;
    ;
    ;
    ;
    ;

    JEL classification:

    • E31 - Macroeconomics and Monetary Economics - - Prices, Business Fluctuations, and Cycles - - - Price Level; Inflation; Deflation
    • C45 - Mathematical and Quantitative Methods - - Econometric and Statistical Methods: Special Topics - - - Neural Networks and Related Topics
    • C49 - Mathematical and Quantitative Methods - - Econometric and Statistical Methods: Special Topics - - - Other
    • C53 - Mathematical and Quantitative Methods - - Econometric Modeling - - - Forecasting and Prediction Models; Simulation Methods

    Statistics

    Access and download statistics

    Corrections

    All material on this site has been provided by the respective publishers and authors. You can help correct errors and omissions. When requesting a correction, please mention this item's handle: RePEc:apk:doctra:2002. See general information about how to correct material in RePEc.

    If you have authored this item and are not yet registered with RePEc, we encourage you to do it here. This allows to link your profile to this item. It also allows you to accept potential citations to this item that we are uncertain about.

    We have no bibliographic references for this item. You can help adding them by using this form .

    If you know of missing items citing this one, you can help us creating those links by adding the relevant references in the same way as above, for each refering item. If you are a registered author of this item, you may also want to check the "citations" tab in your RePEc Author Service profile, as there may be some citations waiting for confirmation.

    For technical questions regarding this item, or to correct its authors, title, abstract, bibliographic or download information, contact: Departamento de Investigación Económica (email available below). General contact details of provider: https://edirc.repec.org/data/bccrrcr.html .

    Please note that corrections may take a couple of weeks to filter through the various RePEc services.

    IDEAS is a RePEc service. RePEc uses bibliographic data supplied by the respective publishers.