Forecasting Monthly Inflation: A MIDAS Regression Application for Turkey

Abstract

Forecasting the short-term price movements is especially important in terms of developing adequate monetary policies during inflationary periods. For countries such as Turkey where inflation targeting policy were adopted and relatively high inflation rates are observed, making short term forecasting using daily data will allow decision processes to react more rapidly. In Turkey, several methods are used by the Central Bank and academicians for estimating the inflation rate. However, in all these methods, covariates are used from the same frequency (mostly monthly) in modelling the inflation rate. In this study, it has been tried to develop a model which can be used in the forecasting of inflation rate by using MIDAS method which allows the series to be used in the same regression equation from different frequency. In the set regression equation, commercial credit interest rate (weekly), TL / US Dollar parity (daily), gold gram price (daily) and oil price (daily) data are used as variables which have the potential to determine the monthly producer price level (PPI) by increasing the input costs. Considering the AIC and SIC criteria, it was found that the best performing model out of four alternatives was the weighted equation according to the Almon polynomial distributed lags method. The in-sample predictive success of the model was found satisfactory.

Keywords

• Inflation forecasting
• Mixed data sampling
• MIDAS
• Turkey

References

1. [1] Alper, C. E., Fendoğlu, S. and Saltoğlu, B. MIDAS volatility forecast performance under market stress: Evidence from emerging stock markets, Economics Letters, 117 (2012) 528-532
2. [2] Armesto, M., Engememann, K. and Owyang, M. T. Forecasting with mixed frequencies. Federal Reserve Bank of St.Louis Review, 92 (2010) 521-536.
3. [3] Barsoum, F. and Stankiewicz, S. Forecasting GDP Growth Using Mixed-Frequency Models with Switching Regimes, University of Konstanz, Working Paper, No. 2013-10. (2013)
4. [4] Baumeister, C., Guérin, P. and Kilian, L. Do High-Frequency Financial Data Help Forecast Oil Prices? The MIDAS Touch, Bank of Canada, Working Paper, No. 2014-11. (2014)
5. [5] Bilgin, D., Jankovic, D. and Lam, A. MIDAS regression using inflation and unemployment to predict GDP, Mimeo. (2018)
6. [6] Breitung, J. and Roling, C. Forecasting inflation rates using daily data: A non-parametric MIDAS approach, Journal of Forecasting, 34 (2015) 588-603.
7. [7] Chen, X. and Ghysels, E. News –good or bad– and its impact on predicting future volatility, Review of Financial Studies, 24 (1) (2011) 46-81.
8. [8] Clements, M. P. and Galvão, A. B. Macroeconomic forecasting with mixed-frequency data: forecasting output growth in the United States, Journal of Business and Economic Statistics, 26 (2008) 546-54.

9. [9] Foroni, C. and Marcellino, M. A Survey of Econometric Methods for Mixed-frequency Data, Norges Bank, Working Paper, 2013-06. (2013)
10. [10] Galvão, A. B. Change in Predictive Ability with Mixed Frequency Data, Queen Mary University, Working Paper, No. 595. (2007)
11. [11] Ghysels, E., Santa-Clara, P. and Valkanov, R. The MIDAS touch: mixed data sampling regressions, Draft paper. (2002)
12. [12] Ghysels, E., Santa-Clara, P. and Valkanov, R. There is a risk-return trade-off after all, Journal of Financial Economics, 76 (2005) 509-548.
13. [13] Ghysels, E., Santa-Clara, P. and Valkanov, R. Predicting volatility: Getting the most out of return data sampled at different frequencies, Journal of Econometrics, 131 (2006) 59-95.
14. [14] Ghysels, E., A. Sinko, and R. Valkanov MIDAS regressions: Further results and new directions, Econometric Reviews, 26 (2007) 53-90.
15. [15] Guerin, P. and M. Marcellino Markov-switching MIDAS Models, CEPR Discussion Papers, 8234. (2011)
16. [16] Guliyev, H. Karma Frekanslı Verilerde MIDAS Regresyon Modellerinin Uygulanması: Türkiye’nin Ekonomik Büyüme Tahmini, Master Thesis, Akdeniz University – ISS, Antalya. (2018)
17. [17] Hoang, D. H. The Effects of Macroeconomic Variables on Asian Stock Market Volatility: A GARCH MIDAS Approach, Master Thesis, Lund University, Sweeden. (2015)
18. [18] Karamelikli, Ö. and Özbilgin, M. Mixed in New Zealand: Nowcasting Labour Markets with MIDAS, Reserve Bank of New Zealand, Analytical Note, No. AN2019/04. (2019)
19. [19] Kingnetr, N., Tungtrakul, T. and Sriboonchitta, S. Forecasting GDP growth in Thailand with different leading indicators using MIDAS regression models, in Robustness in Econometrics, Ed.: V. Kreinovich et al., Springer International Publishing. (2017)
20. [20] Kotze, K. L. Forecasting inflation with high frequency asset price data, University of Stellenbosch, Working Paper. (2005)
21. [21] Kuzin, V., Marcellino, M. and Schumacher, C. MIDAS versus mixed-frequency VAR: Nowcasting GDP in the euro area, Discussion Paper No. 07/2009, Deutsche Bundesbank. (2009)
22. [22] Lebouef, M. and Morel, L. Forecasting Short-term Real GDP Growth in the Euro Area and Japan Using Unristricted MIDAS Regression, Bank of Canada, Working Paper, No. 2014-3. (2014)
23. [23] Libonatti, L. MIDAS Modeling for Core Inflation Forecasting, IDB Working Paper Series, No. IDB-WP-897. (2018)
24. [24] Marcellino, M. Some consequences of temporal aggregation in empirical analysis. Journal of Business & Economic Statistics, 17 (1999) 129-136.
25. [25] Marsilli, C. Nowcasting US inflation using a MIDAS augmented Phillips curve, International Journal of Computational Economics and Econometrics, 7 (2017) 64-77.
26. [26] Mikosch, H. and Solanko, L. Forecasting quarterly Russian GDP growth with mixed-frequency data, Russian Journal of Money and Finance, 78 (2019) 19-35.
27. [27] Modugno, M. Nowcasting Inflation Using High Frequency Data, European Central Bank, Working Paper Series, No. 1324. (2011)
28. [28] Monteforte, L. and Moretti, G. Real-time forecasts of inflation: The role of financial variables, Journal of Forecasting, 32 (2012) 51-61.
29. [29] Ribon, S. and Suhoy, T. Forecasting Short-run Inflation Using Mixed Frequency Data (MIDAS), Bank of Israel, Occasional Paper, 2011.02. (2011)
30. [30] Salisu, A. A. and Ogbonna, A. E. Improving the predictive ability of oil for inflation: An ADL-MIDAS Approach, Centre for Econometric and Allied Research, University of Ibadan Working Papers Series, CWPS 0025. (2017)
31. [31] Schumacher, C. and Breitung, J. Real-time forecasting of German GDP based on a large factor model with monthly and quarterly data, International Journal of Forecasting, 24 (2008) 386–398.
32. [32] Suhoy, T. Monthly Assessments of Private Consumption, Bank of Israel, Discussion Paper, No. 2010-09. (2010)
33. [33] Tay, A. S. Mixing Frequencies: Stock Returns as a Predictor of Real Output Growth, Singapur Management University, Working Paper, No. 34-2006. (2006)
34. [34] Yamak, N., Samut, S. and Koçak, S. Farklı frekanslı seriler altında ekonomik büyüme oranının tahmini, Ekonomi Bilimleri Dergisi, 10 (2018) 34-49.
35. [35] Zhao, X., Han, M., Ding, L. and Calin, A. C. Forecasting carbon dioxide emissions based on a hybrid of mixed data sampling regression model and back propagation neural network in the USA, Environmental Sciences and Pollution Research, 25 (2018) 2899-2910.