Author :
Abstract
Makalemizde (2005: 01-2020: 05) Türkiye ekonomisinin enflasyon ve enflasyon belirsizliği arasındaki ilişkiyi ARMA-GARCH modelini kullanarak araştırdık. Bu konuda yapılmış birçok teorik ve metodolojik çalışma bulunmaktadır. Bazı ampirik çalışmalar enflasyonun enflasyon dalgalanması üzerindeki olumlu etkisine ilişkin Friedman-Ball hipotezini desteklese de, bazıları Cukierman & Meltzer'in enflasyon belirsizliğinin daha yüksek enflasyona katkıda bulunduğuna ilişkin hipotezini desteklemektedir ve son olarak bulgularının geri kalanı, belirsizliği gösteren Hollanda hipotezini doğrulamaktadır. potansiyel enflasyonu düşürmek için tetikler. Çalışmamızda enflasyon, Friedman-Ball hipotezini destekleyen dört-beş gecikme seçimi koşulunun hepsinde belirsizlikte artışa neden olur. Enflasyon belirsizliği, Cukierman & Meltzer hipotezini ve Holland hipotezini desteklemeyen enflasyonda bir artışa neden olmuyor. Bu nedenle Türkiye Cumhuriyet Merkez Bankası'nın enflasyon belirsizliğini dengelemek için enflasyon hedef politikasını başarıyla uygulayarak yanıt vermesi gerektiği sonucuna varılmıştır.
Keywords
Abstract
In our paper, we research the relationship between inflation and inflation uncertainty of Turkish economy for the period of (2005:01-2020:05) using ARMA-GARCH model. There are several theoretical and methodological studies conducted in this topic. Although some of the empirical studies support the Friedman-Ball hypothesis of positive effect of inflation-on-inflation volatility, some of them support Cukierman & Meltzer hypothesis that inflation uncertainty contributes to higher inflation, and finally rest of their findings corroborate Holland hypothesis indicating that uncertainty trigger to decrease potential inflation. In our study inflation causes, an increase in uncertainty in all four five conditions of lag selection, which supports the Friedman-Ball hypothesis. Inflation uncertainty does not cause an increase in inflation, which does not support the Cukierman & Meltzer hypothesis and Holland hypothesis. Therefore, it is concluded that The Central Bank of the Republic of Turkey should respond to stabilize the inflation uncertainty by implementing the inflation targeting policy successfully.
Keywords
- Empirically we can review all studies in the literature into three categories : (1) supporting Friedman-Ball hypothesis (inflation causes of uncertainty)a, (2) Cukierman & Meltzer Hypothesis (1986)b, (Uncertianty Causes Positively Inflation), (3) Holland Hypothesis (1995), (uncertainty lower the average inflation)c.
- Authors Method Period Result Elton, Atsuyuki and Benito (2005)
- APGARCH 1957-2004
- Erkam (2008) ARCH, GARCH and PARCH 1982–2008 Support the Friedman-Ball Fisunoğlu and Özdemir
- GARCH 1987-2003
- and Cukierman and Meltzer Fountas(2001) GARCH 1885-1998 Support the Friedman-Ball Kim( 1993) ARCH 1958–1990
- GARCH-M models 1972-2002 Orhan and Keske (2010) GARCH-M 1984-2005
- hypothesis, and Holland Telatar and Telatar(2003) Granger 1995–2000 Support the Friedman-Ball Thornton (2007) GARCH model
- Wilson (2006)
- bivariate EGARCH-M model 1957 – 2002
- Riaz and Munir (2020) ARMA-GARCH 1998-2018
- Cukierman – Meltzer Farzinvash , Elaahi , Kiaalhosseini & Haashemi Dizaj ( 2016)
- MS – VAR 1990-2015 Support the Friedman-Ball Jiranyakul( 2020) EGARCH 1979 – 2019
- Support Friedman-Ball hypothesis and Cukierman – 3. Data and Methodology In this study, we investigated the relationship between inflation and inflation uncertainty for Turkey over the period of 2005:M1-2020M5 for the inflation, we used the Consumer Price Index (C.P.I.) as a monthly percentage change obtained from The Central Bank of Republic of Turkey (TCMB). In order to attain the index of inflation uncertainty we will use the conditional variance of inflation through GARCH model, the method that we apply for our analysis is ARMA –EGARCH model, and then we investigate causality between two variables with Granger-causilty test. 3. 1. The Empirical Results Table 2 summarizes the descriptive statistics of the inflation variable, CPI, from Turkey over the period of 2005-2020. Table 2. Summary Statistics for CPI Mean 0.762270 Skewness 1.423253 Median 0.650000 Kurtosis 10.12803 Maximum 6.300000 Jarque-Bera 454.109 Minimum -1.44 Sum 141.0200 Std. Dev. 0.892523 Sum Sq. Dev. 146.5738 The Table.2 shows the mean, median, standard deviation (std), minimum value (min), and maximum value (max) are 0.76, 0.65, 6.3, -1.44, 0.89 respectively, for the variable. Prior to fitting the ARIMA ARCH model, series should be stationary, otherwise it should be converted into stationary series. Table 3. Unit Root Test Results for CPI ADF
- The results of the test for the monthly inflation rates in Turkey can be seen in the Table.4 above. In the model AR (12,14) and moving average (1,2,4) satisfy the stationary conditions. Table 5. Heteroskedasticity Test: ARCH F-statistic 0.019274 Prob.F(1,158) 0.8898 Obs*R-squared 0.019515 Prob. Chi-Square (1) 0.8889 F-statistic 1.131880 Prob. F(1,158) 0.8898 Obs*R-squared 4.541186 Prob. Chi-Square (1) 0.8889 F-statistic 1.153394 Prob. F(8,144) 0.3316
- MA(1) 0.270886 0.103386 2.620146 0.0088
- Sample 2007M01 2020M05
- In this study, inflation uncertainty is firstly proxied by the GARCH model, and then the conditional variance based on this estimation is used to as proxy for the inflation. Later, we proceeded to Granger Causality test between two variables. As per the findings, there is a limited degree of persistence of volatility in response to inflationary shocks. Moreover, the Granger causality test conducted indicated that high inflation Granger causes inflation uncertainty, which strongly agrees with the Friedman (1977)and Ball (1992) postulations as cited in (Ran, Zheng-Zheng, Xiao-Lin, and SU, 2018: 41). There was no sufficient evidence in support of the Cukierman-Meltzer hypothesis, as revealed by the causality test results.
- Kim, C.-J. (1993). Unobserved-Component Time Series Models With Markov-Switching Heteroscedasticity: Changes in
- Statistics, 11(3), 341-349. Kontonikas, A. (2004). Inflation and inflation uncertainty in the United Kingdom, evidence from GARCH modelling.
- Economic Modelling, 21(3), 525 - 543. Munir, K., & Riaz,N. (2020). Dynamics of inflation and inflation uncertainty in Pakistan. International Journal of
- Monetary Economics and Finance, 13(2), 130-145. Payne, J. (2008). Inflation and inflation uncertainty: evidence from the Caribbean region. Journal of Economic Studies,
- 35(6), 501-511. Pindyck, R. (1991). Irreversibility, uncertainty, and investment. Journal of Economic Literature, 29, 1110–1148. Ran, T., Zheng-Zheng, L., Xiao-Lin , L ., & Chi-Wei, S. (2018). A Reexamination of Friedmanball’s Hypothesis In Slovakia
- Evidence From Wavelet Analysis. Romanian Journal of Economic Forecasting, 21(5), 41-54. Telatar, F., and Telatar, E. (2003). The relationship between inflation and different sources of inflation uncertainty in
- Turkey. Applied Economics Letters, 10(7), 431-435. Thornton, J. (2007). The Relationship between Inflation and Inflation Uncertainty in Emerging Market Economies.
- 73(4), 858-870. Wilson, B. K. (2006). The links between inflation, inflation uncertainty and output growth: New time series evidence
