Bayesian combined forecasts and Monte Carlo simulations to improve inflation rate predictions in Romania

• Authors: Mihaela Simionescu
• Languages of publication: EN

Abstracts

EN

In this paper we applied the regression approach and Bayesian inference to obtain more accurate forecasts of the inflation rate in the case of the Romanian economy. The necessity of using the most accurate forecasts for the inflation rate is required by the realisation of economic criteria for the accession to the eurozone and by the inflation targeting strategy of the National Bank of Romania. Considering the assumption that simple econometric models provide better forecasts than complex models, in this paper we combined various forecasts from individual models using as prior information the expectations of experts. The empirical findings for Romanian inflation rate forecasts over the horizon of 2016-2018 indicated that a fixed effects model performed better than other simple models (autoregressive moving average model, dynamic model, simple and multiple linear model, VAR, Bayesian VAR, simultaneous equations model). The Bayesian combined forecasts that used experts’ predictions as priors, with a shrinkage parameter tending to infinity, improved the accuracy of all predictions using individual models, outperforming also naïve forecasts and zero and equal weights forecasts. However, predictions based on Monte Carlo simulation outperformed all the scenarios in terms of the mean error and mean absolute error.  

Keywords

EN

forecasts accuracy; Bayesian forecasts combination; shrinkage parameter; econometric model

• Publisher: Wydział Ekonomii Uniwersytetu Ekonomicznego w Poznaniu
• Journal: Research Papers in Economics and Finance
• Year: 2020
• Volume: 4
• Issue: 1
• Pages: 7-20

Dates

published

2020-03-31

Contributors

author

Mihaela Simionescu

References

• Armstrong, J. S. (2005). The Forecasting Canon: Nine Generalizations to Improve Forecast Accuracy. International Journal of Applied Forecasting, 1, 29-35. Available at SSRN: https://ssrn.com/abstract=868496
• Atkeson, A., & Ohanian, L. E. (2001). Are Phillips curves useful for forecasting inflation? Federal Reserve Bank of Minneapolis. Quarterly Review-Federal Reserve Bank of Minneapolis, 25(1), 2-11.
• Baciu, I. C. (2015). Stochastic Models for Forecasting Inflation Rate. Empirical Evidence from Romania. Procedia Economics and Finance, 20, 44-52.
• Budina, N., Maliszewski, W., De Menil, G., &Turlea, G. (2006). Money, inflation and output in Romania, 1992–2000. Journal of International Money and Finance, 25(2), 330-347.
• Coulson, N., Robins, R. (1993). Forecast combination in a dynamic setting, Journal of Forecasting, 12, 63-67.
• Diebold, F.X., Pauly, P. (1990). The use of prior information in forecast combination. International Journal of Forecasting, 6, 503-508.
• Dobrescu, E. (2009). Measuring the Interaction of Structural Changes with Inflation. Journal for Economic Forecasting, 6(5), 5-99.
• Dobrescu, E. (2017). Modelling an Emergent Economy and Parameter Instability Problem. Journal for Economic Forecasting, 20(2), 5-28.
• Dritsakis, N. (2004). A causal relationship between inflation and productivity: An empirical approach for Romania. American Journal of Applied Sciences, 1(2), 121-128.
• Furtula, S., Durkalić, D., & Simionescu, M. (2018). Testing Phillips Curve for Serbian and Romanian Economy. Romanian Statistical Review, 3(2018), 40-51.
• Geweke, J., Whiteman, C. (2006). Bayesian forecasting, Handbook of economic forecasting, 3-80.
• Gomez, M.I., Gonzalez, E.R., Melo, L.F. (2012). Forecasting food inflation in developing countries with inflation regimes. American Journal Of Agricultural Economics, 94, 153-173.
• Gordon, R. J. (1990). The Phillips curve now and then (No. w3393). National Bureau of Economic Research.
• Gordon, R. J., & Stock, J. H. (1998). Foundations of the Goldilocks economy: supply shocks and the time-varying NAIRU. Brookings papers on economic activity, 1998(2), 297-346.
• Granger, C.W.J., Ramanathan, R. (1984). Improved methods of combining forecasts. Journal of Forecasting, 3, 179–204.
• Julio, J.M., Bratu Simionescu, M. (2013). The evaluation of forecasts uncertainty for rate of inflation using a fan chart. Journal of Economic Computation and Economic Cybernetics and Research, 2, 115-128.
• Koop, G., Korobilis, D. (2010).Bayesian Multivariate Time Series Methods for Empirical Macroeconomics, http://personal.strath.ac.uk/gary.koop/kk3.pdf.
• Koop, G., Potter, S. (2003). Forecasting in large macroeconomic panels using Bayesian model averaging. Staff report 163, Federal Reserve Bank of New York.
• Liao, Y. (2014), Inflation Forecast Literature Review 1969-2013, SSRN 2444121, available online at: http://papers.ssrn.com/sol3/papers.cfm?abstract_id=2444121.
• Mihai, N. D., Cristina, S. M., Lucian, B., Lucia, C., & Gabriel, I. (2016). Inflation in Romania and its evolution in view of accession to the eurozone. Journal of Information Systems & Operations Management, 1.
• Poncela, P., Rodríguez, J., Sánchez-Mangas, R., &Senra, E. (2011). Forecast combination through dimension reduction techniques. International Journal of Forecasting, 27(2), 224-237.
• Simionescu, M. (2014). The Performance of Predictions Based on the Dobrescu Macromodel for the Romanian Economy. Journal for Economic Forecasting, (3), 179-195.
• Sovbetov, Y., & Kaplan, M. (2019). Empirical examination of the stability of expectations-Augmented Phillips Curve for developing and developed countries. Theoretical & Applied Economics, 2(2), 63-78.
• Stock, J. H., & Watson, M. W. (2007). Why has US inflation become harder to forecast? Journal of Money, Credit and Banking, 39(s1), 3-33.
• Terceno, A., Vigier, H. (2011). Economic – Financial Forecasting Model of Business Using Fuzzy Relations. Journal of Economic Computation and Economic Cybernetics and Research, 1, 215-233.
• Tian, J & Zhou, Q 2013, Predictability of technical analysis: a combination approach, The 33th International Symposium on Forecasting 2013, Seoul, Coreea, 23-26 June 2013, http://forecasters.org/isf/submissions/proceedings/
• Timmermann, A. (2006). Forecast combination. Handbook of economic forecasting, 135-196.
• Velandia, L. F. M., Maya, R. A. L., & Villamizar-Villegas, M. (2014). Bayesian Combination for Inflation Forecasts: The Effects of a Prior Based on Central Banks’ Estimates (No. 012323). BANCO DE LA REPÚBLICA.
• Wright, J.H. (2008). Bayesian model averaging and exchange rate forecasts. Journal of Econometrics, 146, 329-341.
• Zellner, A. (1986). On assessing prior distributions and Bayesian regression analysis with g-prior distributions. Journal of Econometrics, 40, 183-202.

Identifiers

DOI

10.18559/ref.2020.1.1