Powerful Trend Function Tests that are Robust to Strong Serial Correlation with an Application to the Prebisch-Singer Hypothesis

In this paper we propose tests for hypothesis regarding the parameters of a the deterministic trend function of a univariate time series. The tests do not require knowledge of the form of serial correlation in the data and they are robust to strong serial correlation. The data can contain a unit root and the tests still have the correct size asymptotically. The tests we analyze are standard heteroskedasticity autocorrelation (HAC) robust tests based on nonparametric kernel variance estimators. We analyze these tests using the small-b asymptotic framework recently proposed by Kiefer and Vogelsang (2002). This analysis allows us to analyze the power properties of the tests with regards to bandwidth and kernel choices. Our analysis shows that among popular kernels, there are specific kernel and bandwidth choices that deliver tests with maximal power within a specific class of tests. We apply the recommended tests to the logarithm of a net barter terms of trade series and we find that this series has a statistically significant negative slope. This finding is consistent with the well known Prebisch-Singer hypothesis. Because our tests are robust to strong serial correlation or a unit root in the data, our results in support of the Prebisch-Singer hypothesis are relatively strong