Pairwise Tests of Purchasing Power Parity
对50个国家1957-2001年的季度数据，对所有可能的双边实际汇率进行单位根检验，估计平稳比例，发现只有足够大的失衡才能拒绝无调整的原假设，线性模型下拒绝率约40%，非线性模型下约60%。
Given nominal exchange rates and price data on N + 1 countries\n indexed by i = 0,1,2,…, N, the standard procedure for\n testing purchasing power parity (PPP) is to apply unit root or\n stationarity tests to N real exchange rates all measured relative to a\n base country, 0, often taken to be the U.S. Such a procedure is sensitive\n to the choice of base country, ignores the information in all the other\n cross-rates and is subject to a high degree of cross-section dependence\n which has adverse effects on estimation and inference. In this article, we\n conduct a variety of unit root tests on all possible N(N + 1)/2\n real rates between pairs of the N + 1 countries and estimate the\n proportion of the pairs that are stationary. This proportion can be\n consistently estimated even in the presence of cross-section dependence.\n We estimate this proportion using quarterly data on the real exchange rate\n for 50 countries over the period 1957-2001. The main substantive\n conclusion is that to reject the null of no adjustment to PPP requires\n sufficiently large disequilibria to move the real rate out of the band of\n inaction set by trade costs. In such cases, one can reject the null of no\n adjustment to PPP up to 90% of the time as compared to around 40% in the\n whole sample using a linear alternative and almost 60% using a nonlinear\n alternative.