Türkiye øçin Yeni Bir Enflasyon Göstergesi: Enflasyon Aktivite Endeksi

Enflasyon oranlarÕ, merkez bankalarÕnÕn para politikalarÕnda ataca÷Õ adÕmlara temel teúkil etmesi sebebiyle finansal piyasa katÕlÕmcÕlarÕ tarafÕndan yakÕndan takip edilmektedir. Bu çalÕúmada, ilk olarak enflasyonu etkileme potansiyeline sahip 252 makro ve finansal de÷iúkenden oluúan bir veri seti kullanÕlarak dinamik faktör modeli ile yeni bir enflasyon göstergesi niteli÷i taúÕyan enflasyon aktivite endeksi geliútirilmiútir. økinci aúamada ise, enflasyon aktivite endeksinin örneklem dÕúÕ enflasyon tahmin performansÕ, enflasyonun gecikmeli de÷erlerinden oluúan oto-regresif model (AR) ile karúÕlaútÕrÕlmÕútÕr. Örneklem dÕúÕ tahminler geniúleyen pencere yönteminin yanÕ sÕra kayan pencerelerde de tekrarlanarak enflasyondaki kÕrÕlmalar ve trend de÷iúimleri dikkate alÕnmÕútÕr. Elde edilen bulgular, makro ve finansal de÷iúkenleri dikkate alan büyük veri setinden oluúturulan enflasyon aktivite endeksinin enflasyonu tahmin etmede ek bir bilgi sundu÷unu göstermektedir. Özellikle, enflasyon aktivite endeksini içeren modellerin örneklem dÕúÕ uzun vadeli tahminlerinde baz modele kÕyasla hata oranlarÕnÕn ciddi úekilde azaldÕ÷Õ görülmektedir. Bu çerçevede, türetilen endeksin politika yapÕcÕlar ve piyasa katÕlÕmcÕlarÕ için yeni bir enflasyon göstergesi olarak fayda sa÷layaca÷Õ de÷erlendirilmektedir. ÇalÕúmada sunulan de÷erlendirmeler tamamÕyla yazarlara aittir; TCMB’nin resmi görüúleri olarak yorumlanmamalÕdÕr.

A New Inflation Gauge for Turkey: Inflation Activity Index

Financial market institutions closely monitor inflation rates since central banks regard inflation as a basic input for their monetary policy decisions. In this study, as a first step, a new inflation indicator is developed utilizing a dynamic factor model based on a data set consisting of 252 macro and financial variables that have potential to influence inflation. In the following step, out-of-sample forecast performance of the inflation activity index is compared with the auto-regressive (AR) model that consists of the lagged values of inflation. Out-of-sample forecasting exercise is utilized with expanding and rolling window strategies to take into account the breakpoints and trend shifts. The findings indicate that the inflation activity index, which is generated from the large data set, provides additional information for out-of-sample forecasting of inflation rates. In particular, results indicate that forecast errors decline significantly in the long-term forecast horizon compared to the baseline model. Overall, the newly constructed index is assessed to be an new inflation gauge for policymakers and market participants.

___

  • Akdeniz, C. ve ÇatÕk, A.N. (2017). Türkiye için Finansal KoúullarÕn Bir Analizi: Faktör ve VAR Modellerinden Bulgular”, Eskiúehir Osmangazi Üniversitesi øøBF Dergisi, 12(1), 99-120. (http://dergipark.gov.tr/download/article- file/284601).
  • Amstad M. ve Fischer A. (2009). Are Weekly Inflation Forecasts Informative?, Oxford Bulletin of Economics and Statistics, Vol.71, Issue 2(04).
  • Amstad, M. ve Simon Potter (2009). Real-time Underlying inflation gauge for Monetary Policy Makers, FRBNY Staff Report, No. 420.
  • Amstad M., Huan, Y., ve Ma, G. (2014). Developing an underlying inflation gauge for China. BIS Working Papers, No. 465.
  • Amstad M., Simon Potter ve Robert Rich (2014): “The FRBNY staff Underlying Inflation Gauge (UIG)”, FRBNY Staff Report, No. 671.
  • Bai, J., (2003). Inferential Theory For Factor Models of Large Dimensions. Econometrica, 71(1), pp.135-171.
  • Bragoli, D. ve Modugno, M., (2017). A now-casting model for Canada: Do US Variables Matter?. International Journal of Forecasting, 33(4), 786-800.
  • Claus, I. (2000). Is the Output Gap A Useful øndicator Of ønflation. Reserve Bank of New Zealand Discussion Paper Series, DP2000/05.
  • Engle, R. F. ve M. W. Watson (1981). A One-Factor Multivariate Time Series Model Of Metropolitan Wage Rates. Journal of the American Statistical Association 76 (376), 774-781.
  • Forni, M., Hallin, M., Lippi, M. ve Reichlin, L., (2000). The Generalized Dynamic-Factor Model: Identification And Estimation. Review of Economics and statistics, 82(4), 540-554.
  • Geweke, J., (1977). The Dynamic Factor Analysis Of Economic Time Series. Latent Variables øn Socio-Economic Models.
  • Giannone, D., Reichlin, L., ve Small, D. (2008). Nowcasting: The Real-Time ønformational Content Of Macroeconomic Data. Journal of Monetary Economics, 55(4), 665-676.
  • Modugno, M., (2013). Now-Casting ønflation Using High Frequency Data. International Journal of Forecasting, 29(4), 664-675.
  • Modugno, M., Soybilgen, B. ve Yazgan, E., (2016). Nowcasting Turkish GDP and News Decomposition. International Journal of Forecasting, 32(4), 1369-1384.
  • Moosa, I. A. (1998). Are Commodity Prices A Leading øndicator Of ønflation?. Journal of Policy Modeling, 20(2), 201-212.
  • Quah, D. ve Sargent, T.J., (1993). A Dynamic Index Model For Large Cross Sections. In Business cycles, indicators and forecasting (pp. 285-310). University of Chicago Press.
  • Sargent, T.J. ve Sims, C.A., (1977). Business Cycle Modeling Without Pretending To Have Too Much A Priori Economic Theory. New Methods øn Business Cycle Research, 1, pp.145-168.
  • Stock, J.H. ve Watson, M.W., (1989). New Indexes of Coincident and Leading Economic øndicators. NBER Macroeconomics Annual, 4, pp.351-394.
  • Stock, J. H., ve Watson, M. W. (1999). Forecasting inflation. Journal of Monetary Economics, 44(2), 293-335.
  • Stock, J. H., ve Watson, M. W. (2002). Macroeconomic Forecasting Using Diffusion Indexes. Journal of Business & Economic Statistics, 20(2), 147-162.