We show that the use of the real effective exchange rate to test for purchasing power parity, as in Astorga (2012) and other studies, is subject to a problem that biases tests against finding evidence of PPP. The problem is illustrated using Astorga´s data on six Latin American countries.

We study whether there is a long-run relationship between Mexican current account (CA) revenues and expenditures. Our results show that evidence in favor of this claim is drawn only when (at least) three structural break levels are allowed. The CA therefore behaves as a broken-mean stationary...

Mexican economic historiography recognizes the key role that public investment played in the country's economic performance from the post-revolutionary period until the beginning of the economic liberalization that began in the mid-1980s. However, there is no concrete empirical evidence that...

Im et al. (Unpublished working paper, <CitationRef CitationID="CR11">2008</CitationRef>) develop cointegration tests using stationary instrumental variables. Their tests avoid the need to simulate critical values for the cointegration estimations, especially problematic in the presence of a nuisance parameter. Likewise, bootstrapping...</citationref>

Three well-known single equation cointegration tests are employed to test for purchasing power parity (PPP) in updated version of the data set developed by Taylor (2002). Results of the tests differ somewhat. The Engle-Granger two-step procedure indicates substantial support for PPP with respect...

In recent work Im, Lee, and Enders (2006) use stationary instrumental variables to test for cointegrating relationships. The advantage of their approach is that the t-statistics are asymptotically standard normal and the familiar critical values of the normal distribution may be used to assess...

It has been found that the t-statistic for testing the null of no relationship between two independent variables diverges asymptotically under a wide variety of nonstationary data generating processes. This paper introduces a simple method which guarantees convergence of this t-statistic to a...

It has been found that the t-statistic for testing the null of no relationship between two independent variables diverges asymptotically under a wide variety of nonstationary data generating processes. This paper introduces a simple method which guarantees convergence of this t-statistic to a...