Research Article
BibTex RIS Cite

MIDAS ve MF-VAR Modelleri ile GSYH Ön Tahmini

Year 2021, Volume: 11 Issue: 1, 1 - 22, 31.07.2021

Abstract

Küreselleşen dünya ekonomisi ve yaşanan teknolojik gelişmeler, ekonominin konjonktürel durumunun tespiti doğrultusunda uygun ekonomi politikalarının olabildiğince erken zamanda üretilme zorunluluğunu ortaya çıkarmıştır. Bu amaçla Eurostat öncülüğünde başlatılan çalışmaların başında, ekonominin mevcut durumu hakkında bilgi sağlayan temel göstergelerden olan GSYH ön tahmin çalışmaları yer almaktadır. Ön tahmin uygulamaları ile üç aylık GSYH'nin eldeki mevcut veriler kullanılarak ekonometrik modeller aracılığıyla nihai tahmin döneminden daha erken zamanda hesaplanmasına imkan sağlanmıştır. Bu çalışmada da GSYH çeyreklik büyüme oranının referans dönemin sona ermesinden 45 gün sonra elde edilmesine yönelik Türkiye uygulaması gerçekleştirilmiştir.



t+45 anında GSYH'nin ön tahmininin hesaplanma aşamasında ilk olarak, iktisadi teori çerçevesinde GSYH ile ilişkili 28 tane gösterge belirlenerek göstergelerin zaman serisi özellikleri incelenmiştir. Ön tahmin hesabında, farklı frekanslı verilerde yer alan tüm bilgiyi kullanarak aynı anda modellenmesine olanak sağlayan Almon Polinomlu MIDAS regresyon modelleri ile göstergelerin dinamik etkilerinin denklem sisteminde incelendiği MF-VAR modelleri kullanılmıştır. Belirtilen iki farklı modelden ön tahminler elde edilmiş olup modellerin karşılaştırmalı analizi gerçekleştirilmiştir. Tahmin uzunluklarının tahmin performansına etkisini de değerlendirmek amacıyla örneklem dışı tahminlerde 1 yıllık süreyi kapsayan 4 çeyrek dönem için tahminler elde edilerek RMSE değerleri incelenmiştir. Sonuç olarak kısa ve uzun dönem tahminlerinde MIDAS modellerinin MF-VAR modellerinden daha iyi performansa sahip olduğu ileri sürülebilir.

References

  • Alper, C. E., Fendoğlu, S. ve Saltoğlu, B. (2012). MIDAS Volatility Forecast Performance Under Market Stress: Evidence From Emerging Stock Markets. Economics Letters, 117(2): 528-532.
  • Al-Qawasmı, M. (2014). Forecasting Palestinian Gross Domestic Product Using Mixed Data Sampling Regression Techniques, Yüksek Lisans Tezi, Birzeit University.
  • Anesti, N., Hayes, S. ve Moreira, A. (2017). Peering Into The Present: The Bank's Approach to GDP Nowcasting, Bank of England Quarterly Bulletin 2017 Q2.
  • Aprigliano, V., Foroni, C., Marcellino, M., Mazzi, G. ve Venditti, F. (2017). A Daily Indicator of Economic Growth for The Euro Area. International Journal of Computational Economics and Econometrics, 7(1-2): 43-63.
  • Armesto, M., Engemann, K. ve Owyang, M. (2010). Forecasting with Mixed Frequencies. Federal Reserve Bank of St. Louis Review, 92(6): 521-36.
  • Berksun, D. (2019). Electricity Consumptıon And Economic Growth In Turkey: An MF-VAR Approach, Yüksek Lisans Tezi, Bilkent Üniversitesi, Ankara.
  • Clements, M. P. ve Galvão, A. (2008). Macroeconomic Forecasting with Mixed-Frequency Data: Forecasting Output Growth in The United States. Journal of Business and Economic Statistics, 26(4): 546-54.
  • Eurostat (2016). Overview Of GDP Flash Estimation Methods. Publications Office of The European Union, Luxembourg.
  • Ferrara L. ve Marsilli, C. (2014). Nowcasting Global Economic Growth: A Factor-Augmented Mixed-Frequency Approach. Working Papers 515, Banque De France.
  • Foroni, C., Marcellino, M. ve Schumacher, C. (2015). Unrestricted Mixed Data Sampling (MIDAS): MIDAS Regressions With Unrestricted Lag Polynomials. Journal of The Royal Statistical Society: Series A (Statistics In Society), 178(1): 57-82.
  • Ghysels, E., Santa-Clara, P. ve Valkanov, R. (2004). The MIDAS Touch: Mixed Data Sampling Regression Models. California Digital Library, University Of California.
  • Ghysels, E., Sinko, A. ve Valkanov, R. (2007). MIDAS Regressions: Further Results And New Directions. Econometric Reviews, 26(1): 53-90.
  • Ghysels, E. ve Andreou, E. (2009). Structural Breaks in Financial Time Series. Handbook of Financial Time Series, Springer, Berlin.
  • Ghysels, E. ve Kourtellos, A. (2010). Forecasting with Mixed-Frequency Data. Oxford Journal of Economic Forecasting, 158(1): 12.
  • Ghysels, E., Hill, J. ve Motegi, K. (2015). Simple Granger Causality Tests for Mixed Frequency Data. Tohoku University.
  • Ghysels, E. (2016). Macroeconomics and The Reality of Mixed Frequency Data. Journal of Econometrics, 193(2): 294-314.
  • Ghysels, E., Hill, J. ve Motegi, K. (2016). Testing for Granger Causality with Mixed Frequency Data. Journal of Econometrics, 192: 207-230.
  • Ghysels, E. ve Marcellino, M. (2018). Applied Economic Forecasting Using Time Series Methods. Oxford University Press, United States of America.
  • Guliyev, H. (2018). Karma Frekanslı Verilerde MIDAS Regresyon Modellerinin Uygulanması: Türkiye’nin Ekonomik Büyüme Tahmini. Yüksek Lisans Tezi, Akdeniz Üniversitesi, Antalya.
  • Günay, M. (2018). Nowcasting Annual Turkish GDP Growth With MIDAS. TCMB Ekonomi Notu.
  • INE, Quarterly Spanish National Accounts, Press Release 2019Q2, Methodological Note, https://www.ine.es/en/daco/daco42/daco4214/cntr0219a_en.pdf [Ziyaret Tarihi: 15 Ağustos 2019]
  • Kuzin, V., Marcellino, M. ve Schumacher, C. (2009). Pooling Versus Model Selection For Nowcasting with Many Predictors: An Application To German GDP. Deutsche Bundesbank Discussion Paper, Series 1: Economic Studies.
  • Litterman R.B. (1983). A Random Walk, Markov Model for The Distribution of Time Series, Journal of Business And Economic Statistics, 1: 169-173.
  • Marcellino, M. (1999). Some Consequences of Temporal Aggregation in Empirical Analysis. Journal of Business & Economic Statistics, 17(1): 129-136.
  • Marsilli, C. (2017). Nowcasting US Inflation Using A MIDAS Augmented Phillips Curve. International Journal of Computational Economics And Econometrics, 7(1-2): 64-77.
  • Schorfheide, F. ve Song, D. (2012). Real-Time Forecasting with a Mixed-Frequency VAR. Federal Reserve Bank of Minneapolis Research Department, Working Paper 701.
  • Schumacher, C. (2014). MIDAS and Bridge Equations. Discussion Paper, Deutsche Bundesbank, No 26, Frankfurt.
  • Sims, C. ve Zha, T. (1998). Bayesian Methods For Dynamic Multivariate Models. International Economic Review, 39(4): 949–968.
  • TCMB, https://evds2.tcmb.gov.tr/ [Ziyaret Tarihi: 1 Haziran 2021]
  • TÜİK, http://www.tuik.gov.tr/ [Ziyaret Tarihi: 1 Haziran 2021]
  • Yamak, N. ve Samut, S. (2018). MIDAS Granger Nedensellik Testi (MF-VAR): GSYH ve İşsizlik Oranı. İktisat Seçme Yazılar, Celepler Matbaacılık Yayın Ve Dağıtım, Trabzon.
  • Yamak, N., Samut, S. ve Koçak, S. (2018). Farklı Frekanslı Veriler Altında Ekonomik Büyüme Oranının Tahmini. Ekonomi Bilimleri Dergisi, 10(1): 35-47.

GDP Flash Estimate with MIDAS and Mixed Frequency VAR

Year 2021, Volume: 11 Issue: 1, 1 - 22, 31.07.2021

Abstract

The globalized world economy and the technological developments experienced have revealed the necessity of producing appropriate economic policies as early as possible in line with the determination of the economic situation. For this purpose, the flash estimate studies of GDP, one of the basic indicators that provide information about the current state of the economy, are among the initiatives started under the leadership of Eurostat. With the flash estimate applications, the quarterly GDP has been provided to be calculated earlier than the final estimation period by using econometric models. In this study GDP quarterly growth rate for Turkey at about 45 days after the end of the reference period were carried out. At the stage of calculating the flash estimate of GDP at t+45; 28 related indicators were determined within the framework of economic theory and the time series characteristics of the indicators were examined in the first instance. In the flash estimate calculation, Almon Polynomial MIDAS regression models, which allow simultaneous modeling using all the information in different frequency data, and MF-VAR models in which the dynamic effects of the indicators are examined in the equation system were used. Flash estimates were obtained from the two different models mentioned and comparative analysis of the models was performed. In order to evaluate the effect of forecast lengths on forecast performance, RMSE values were obtained by obtaining forecasts for 4 quarters covering a 1 year period in out-of-sample forecasts. As a result, it can be argued that MIDAS models perform better than MF-VAR models in short and long term estimations.


References

  • Alper, C. E., Fendoğlu, S. ve Saltoğlu, B. (2012). MIDAS Volatility Forecast Performance Under Market Stress: Evidence From Emerging Stock Markets. Economics Letters, 117(2): 528-532.
  • Al-Qawasmı, M. (2014). Forecasting Palestinian Gross Domestic Product Using Mixed Data Sampling Regression Techniques, Yüksek Lisans Tezi, Birzeit University.
  • Anesti, N., Hayes, S. ve Moreira, A. (2017). Peering Into The Present: The Bank's Approach to GDP Nowcasting, Bank of England Quarterly Bulletin 2017 Q2.
  • Aprigliano, V., Foroni, C., Marcellino, M., Mazzi, G. ve Venditti, F. (2017). A Daily Indicator of Economic Growth for The Euro Area. International Journal of Computational Economics and Econometrics, 7(1-2): 43-63.
  • Armesto, M., Engemann, K. ve Owyang, M. (2010). Forecasting with Mixed Frequencies. Federal Reserve Bank of St. Louis Review, 92(6): 521-36.
  • Berksun, D. (2019). Electricity Consumptıon And Economic Growth In Turkey: An MF-VAR Approach, Yüksek Lisans Tezi, Bilkent Üniversitesi, Ankara.
  • Clements, M. P. ve Galvão, A. (2008). Macroeconomic Forecasting with Mixed-Frequency Data: Forecasting Output Growth in The United States. Journal of Business and Economic Statistics, 26(4): 546-54.
  • Eurostat (2016). Overview Of GDP Flash Estimation Methods. Publications Office of The European Union, Luxembourg.
  • Ferrara L. ve Marsilli, C. (2014). Nowcasting Global Economic Growth: A Factor-Augmented Mixed-Frequency Approach. Working Papers 515, Banque De France.
  • Foroni, C., Marcellino, M. ve Schumacher, C. (2015). Unrestricted Mixed Data Sampling (MIDAS): MIDAS Regressions With Unrestricted Lag Polynomials. Journal of The Royal Statistical Society: Series A (Statistics In Society), 178(1): 57-82.
  • Ghysels, E., Santa-Clara, P. ve Valkanov, R. (2004). The MIDAS Touch: Mixed Data Sampling Regression Models. California Digital Library, University Of California.
  • Ghysels, E., Sinko, A. ve Valkanov, R. (2007). MIDAS Regressions: Further Results And New Directions. Econometric Reviews, 26(1): 53-90.
  • Ghysels, E. ve Andreou, E. (2009). Structural Breaks in Financial Time Series. Handbook of Financial Time Series, Springer, Berlin.
  • Ghysels, E. ve Kourtellos, A. (2010). Forecasting with Mixed-Frequency Data. Oxford Journal of Economic Forecasting, 158(1): 12.
  • Ghysels, E., Hill, J. ve Motegi, K. (2015). Simple Granger Causality Tests for Mixed Frequency Data. Tohoku University.
  • Ghysels, E. (2016). Macroeconomics and The Reality of Mixed Frequency Data. Journal of Econometrics, 193(2): 294-314.
  • Ghysels, E., Hill, J. ve Motegi, K. (2016). Testing for Granger Causality with Mixed Frequency Data. Journal of Econometrics, 192: 207-230.
  • Ghysels, E. ve Marcellino, M. (2018). Applied Economic Forecasting Using Time Series Methods. Oxford University Press, United States of America.
  • Guliyev, H. (2018). Karma Frekanslı Verilerde MIDAS Regresyon Modellerinin Uygulanması: Türkiye’nin Ekonomik Büyüme Tahmini. Yüksek Lisans Tezi, Akdeniz Üniversitesi, Antalya.
  • Günay, M. (2018). Nowcasting Annual Turkish GDP Growth With MIDAS. TCMB Ekonomi Notu.
  • INE, Quarterly Spanish National Accounts, Press Release 2019Q2, Methodological Note, https://www.ine.es/en/daco/daco42/daco4214/cntr0219a_en.pdf [Ziyaret Tarihi: 15 Ağustos 2019]
  • Kuzin, V., Marcellino, M. ve Schumacher, C. (2009). Pooling Versus Model Selection For Nowcasting with Many Predictors: An Application To German GDP. Deutsche Bundesbank Discussion Paper, Series 1: Economic Studies.
  • Litterman R.B. (1983). A Random Walk, Markov Model for The Distribution of Time Series, Journal of Business And Economic Statistics, 1: 169-173.
  • Marcellino, M. (1999). Some Consequences of Temporal Aggregation in Empirical Analysis. Journal of Business & Economic Statistics, 17(1): 129-136.
  • Marsilli, C. (2017). Nowcasting US Inflation Using A MIDAS Augmented Phillips Curve. International Journal of Computational Economics And Econometrics, 7(1-2): 64-77.
  • Schorfheide, F. ve Song, D. (2012). Real-Time Forecasting with a Mixed-Frequency VAR. Federal Reserve Bank of Minneapolis Research Department, Working Paper 701.
  • Schumacher, C. (2014). MIDAS and Bridge Equations. Discussion Paper, Deutsche Bundesbank, No 26, Frankfurt.
  • Sims, C. ve Zha, T. (1998). Bayesian Methods For Dynamic Multivariate Models. International Economic Review, 39(4): 949–968.
  • TCMB, https://evds2.tcmb.gov.tr/ [Ziyaret Tarihi: 1 Haziran 2021]
  • TÜİK, http://www.tuik.gov.tr/ [Ziyaret Tarihi: 1 Haziran 2021]
  • Yamak, N. ve Samut, S. (2018). MIDAS Granger Nedensellik Testi (MF-VAR): GSYH ve İşsizlik Oranı. İktisat Seçme Yazılar, Celepler Matbaacılık Yayın Ve Dağıtım, Trabzon.
  • Yamak, N., Samut, S. ve Koçak, S. (2018). Farklı Frekanslı Veriler Altında Ekonomik Büyüme Oranının Tahmini. Ekonomi Bilimleri Dergisi, 10(1): 35-47.
There are 32 citations in total.

Details

Primary Language Turkish
Subjects Economics
Journal Section Research Articles
Authors

Merve Aytekin This is me 0000-0001-9856-2746

Furkan Emirmahmutoglu 0000-0001-7358-3567

Publication Date July 31, 2021
Published in Issue Year 2021 Volume: 11 Issue: 1

Cite

APA Aytekin, M., & Emirmahmutoglu, F. (2021). MIDAS ve MF-VAR Modelleri ile GSYH Ön Tahmini. İstatistik Araştırma Dergisi, 11(1), 1-22.