Published
Real-time Forecasting of Colombian Consumer Price Inflation Using High-Dimensional Methods
Pronósticos en tiempo real de la inflación de los precios al consumidor en Colombia con métodos de alta dimensión
DOI:
https://doi.org/10.15446/rce.v48n2.116414Keywords:
Ensembles, Forecasting inflation, Lasso, Machine learning, Penalized regression, Shrinkage. (en)Aprendizaje de máquinas, Lasso, Métodos de ensamble, Pronósticos de la inflación, Regresión penalizada. (es)
Downloads
In this paper, we examine the effectiveness of high-dimensional methods for real-time forecasting of inflation in Colombia. We utilize statistical dimension reduction techniques, such as sparse principal components and dynamic factor analysis, alongside machine learning algorithms that incorporate shrinkage methods. Our evaluation of out-of-sample forecasts, using a dataset of 102 macroeconomic and financial indicators, indicates that ensembles of multiple underlying models can enhance forecast accuracy for horizons of 11 and 12 months ahead. Additionally, stepwise models are suitable for horizons between 4 and 10 months ahead, and spectral component models are effective for short horizons.
En este documento, se examina la efectividad de los métodos de alta dimensión para pronosticar en tiempo real la inflación en Colombia. Se utilizan técnicas estadísticas de reducción de dimensión, como componentes principales dispersos y análisis factorial dinámico, junto con algoritmos de aprendizaje automático. La evaluación de pronósticos fuera de muestra, utilizando 102 indicadores macroeconómicos y financieros, indica que la síntesis de múltiples modelos subyacentes mejora la precisión de las predicciones para horizontes de 11 y 12 meses. Además, los modelos secuenciales son adecuados para horizontes de entre 4 y 10 meses, y los modelos de componentes espectrales son efectivos para horizontes cortos.
References
Araujo, G. & Gaglianone, W. (2020), 'Machine learning methods for inflation forecasting in brazil: new contenders versus classical models', Central Bank of Barzil Working Paper pp. 1-39.
Both, T. & Nibbering, D. (2020), Subspace methods, Springer, chapter 9, pp. 267-326. In: Macroeconomic Forecasting in the Era of Big Data: Theory and Practice.
Chernozhukov, V., Hansen, C. & Liao, Y. (2017), 'A lava attack on the recovery of sums of dense and sparse signals', Annals of Statistics 45, 37-76.
Elliot, G., Gargano, A. & Timmermann, A. (2015), 'Complete subset regression with large dimensional sets of predictors', Journal of Economics Dynamics and Control 54(54), 86-110.
Fulton, C. & Hubrisch, K. (2020), 'Forecasting us inflation in real time', Finance and Economics Discussion Series 2021-014. Washington: Board of Governors of the Federal Reserve System, https://doi.org/10.17016/FEDS.2021.014. pp. 1-30.
Garcia, M., Medeiros, M. & Vasconcelos, G. (2017), 'Real-time inflation forecasting with high-dimensional models: The case of brazil', International Journal of Forecasting 33(3), 679-693.
Gianone, D., Lenza, W. & Primiceri, G. (2018), 'Economic predictions with big data: The illusion of sparsity', Federal Reserve Bank of New York Sta Reports, no. 847 pp. 1-26.
Hastie, T., Tishbirani, R. & Friedman, J. (2001), The elemnts of statistical learning: data mining, inference, and prediction, 2 edn, Springer.
Inoue, A. & Killian, L. (2008), 'How useful is bagging in forecasting economic time series? a case study of u.s. consumer price inflation', Journal of the American Statistical Association 103(482), 511-522.
Kim, H. & Swanson, N. (2014), 'Forecasting financial and macroeconomic variables using data reduction methods: New empirical evidence', Journal of econometrics (178), 352-367.
Kim, H. & Swanson, N. (2018), 'Mining big data using parsimonous factor, machine learning, variable selection and shrinkage methods', International Journal of Forecasting 34(2), 339-354.
Kock, A., M. M. & Vasconcelos, G. (2020), Penalized Time Series Regression, Springer, chapter 7, pp. 193_277. In: Arango, L. and Hamann, F. ed., (2020). Macroeconomic Forecasting in the Era of Big Data: Theory and Practice.
Marcellino, M., Stock, J. & Watson, M. (2006), 'A comparison of direct and iterated multistep ar methods for forecasting macroeconomic time series', Journal of econometrics 135(135), 499-526.
Peña, D., Tsay, R. & Zamar, R. (2019), 'Empirical dynamic quantiles for visualizating of high-dimensional time series', Technometrics 61, 429-444.
Silva, G. & Gaglianone, W. (2020), 'Machine learning methods for inflation forecasting in brazil. new contenders versus classical models', Journal of Business and Economic Statistics 30(30), 481-493.
Stock, J. & Watson, M. (2012), 'Generalized shrinkage methods for forecasting using many predictors', Journal of Business and Economic Statistics 30(30), 481-493.
Wainwright, M. (2019a), Introduction, Cambridge Series in Statistical and Probabilistic Mathematics, chapter 1, pp. 236_257. In: High-Dimensional Statistics: A Non-symptotic Viewpoint.
Wainwright, M. (2019b), Principal component anlysis in high-dimensions, Cambridge Series in Statistical and Probabilistic Mathematics, chapter 8, pp. 236-257. In: High-Dimensional Statistics: A Non-Asymptotic Viewpoint.
Zou, H., Hastie, T. & Tibshirani, R. (2006), 'Sparse principal component analysis', Journal of Computational and Graphical Statistics 15(2), 265-286.
Dimensions
PlumX
Article abstract page views
Downloads
How to Cite
License
This work is licensed under a Creative Commons Attribution 4.0 International License.
- Authors retain copyright and grant the journal right of first publication with the work simultaneously licensed under a Creative Commons Attribution License that allows others to share the work with an acknowledgement of the work's authorship and initial publication in this journal.
- Authors are able to enter into separate, additional contractual arrangements for the non-exclusive distribution of the journal's published version of the work (e.g., post it to an institutional repository or publish it in a book), with an acknowledgement of its initial publication in this journal.
- Authors are permitted and encouraged to post their work online (e.g., in institutional repositories or on their website) prior to and during the submission process, as it can lead to productive exchanges, as well as earlier and greater citation of published work (See The Effect of Open Access).
