This study uses the standard Augmented Dickey-Fuller (ADF) linear test and a new nonlinear test to model the behavior of Real Effective Exchange Rate (REER) for five middle income economies and an oil-producing country. Using monthly data from 1982M1 to 2006M1, the estimation results indicate that the new nonlinear test is more successful in detecting the existence of multi-country Purchasing Power Parity (PPP) than ADF test.

The study of real exchange rate is important to analyze the competitiveness of a country as well as to test the Purchasing Power Parity (PPP) theory. Apart from other implications, PPP is an important building block in monetary approach to the exchange rate and exchange market pressure. The existence of PPP implies that domestic and international markets for goods and foreign exchange are integrated and the monetary authorities need to be careful in pursuing monetary policy.

As the linear models have been relatively less successful in validating the PPP assumption, there is an increasing trend towards employing nonlinear methodologies to capture the behavior of real exchange rate which, if found stationary, indicates the presence of PPP. A recent survey on the topic can be found in Sarno (2005). Some other important studies on the topic are Michael et al. (1997), Sarno (2000), Taylor et al. (2001), Sarno and Taylor (2002), Liew et al. (2004), Sarno (2005), Bahmani-Oskooee et al. (2007), and Zhou (2008).

Using monthly data, this study examines the behavior of Real Effective Exchanger Rate (REER) for a group of five middle income countries, i.e., Malaysia, Pakistan, the Philippines, Poland and Paraguay, and an oil-producing economy, Saudi Arabia. The data used for estimation cover the period from 1982M1 to 2006M1. Finding the stationarity of REER, as compared to other studies which employ bilateral exchange rate, reflects testing the multi-country version of PPP. If REER is found to be stationary, it implies that PPP exists not only with respect to a country's bilateral trading partner, but also in regard to its many trading partners. To check the stationarity of REER, the study uses Kapetanious, Shin and Snell's (KSS) methodology (Kapetanious et al., 2003). This methodology is more powerful in detecting unit root in a series than the Augmented Dickey-Fuller (ADF) test. The null hypothesis in KSS test is the presence of unit root against the alternative hypothesis of nonlinear stationary Smooth Transition Autoregressive (STAR) process.

Monetary Economics Journal, Nonlinear STAR Framework, Power Parity Theory, International Markets, Nonlinear Methodologies, Conventional Linear Methodologies, Nonlinear Techniques, Null Hypothesis, Monetary Policy, Globally Stationary Process, Domestic Markets.