Testing for Stationarity of Inflation Rates With Covariates

Abstract

This paper distinguishes between the stationarity and nonstationarity of inflation in 18 OECD countries through several unit root tests with covariates. These covariate tests are more powerful than the conventional ones through correlated covariates. Both unit root and stationarity null hypotheses are tested in this study. Our empirical results indicate that the efficient univariate unit root tests fail to reject the unit‐root hypothesis for 15 countries, whereas the covariate tests provide strong evidence in support of mean reversion in inflation for almost all countries. Overall, the findings unveil new evidence in favor of stationarity in international inflation. Policy implications on the empirical results are provided as well.

A Revisit to the Stationarity of Oecd Inflation: Evidence From Panel Unit-Root Tests and the Covariate Point Optimal Test*

“…Recently, Tsong and Lee (2010) carry out a battery of covariate tests proposed by Hansen (1995), Elliott and Jansson (2003), and Jansson (2004) to the quarterly inflation rates in 15 OECD countries, and provide strong evidence supporting mean reversion in these countries. The potential covariates considered in Tsong and Lee (2010) are selected from economic theory, including nominal interest rate, unemployment rate, GDP deflator, and GDP. By doing so, they are able to quantify which potential covariate for a specific covariate test can or cannot deliver adequate evidence in favour of mean‐reverting dynamics in inflation.…”

Bootstrapping Covariate Unit Root Tests: An Application to Inflation Rates

“…The choice of covariates is important since it relates to the achieved power and thus affects the empirical results. In this study, we follow the literature (e.g., Amara and Papell, 2006; Elliott and Pesavento, 2006; Tsong and Lee, 2010), and choose the covariates based on economic models. According to the Fisher hypothesis, inflation is expected to be related to nominal interest rates.…”

“…In a related paper, Lee and Tsong (2009) proposed a bootstrap scheme for the covariate stationarity tests of Jansson (2004), giving strong evidence in support of mean reversion in inflation in most countries of the G‐10. Besides, Tsong and Lee (2010) applied the extant three kinds of unit root tests with covariates to inflation in 15 OECD countries, and found strong evidence against unit‐root dynamics for almost all countries.…”

“…Undoubtedly, the high‐power properties of the CP T tests can significantly improve the likelihood of correctly distinguishing a process with a unit root from a slowly mean reverting process. Accordingly, the CP T tests have been employed in many empirical studies (e.g., Amara and Papell, 2006; Elliott and Pesavento, 2006; Tsong and Lee, 2010).…”