A CROSS-SECTIONAL METHOD FOR RIGHT-TAILED PANIC TESTS UNDER A MODERATELY LOCAL TO UNITY FRAMEWORK
研究了PANIC检验在成分中度爆炸时的右尾版本性质，发现存在尺寸扭曲和非单调功效问题，并提出一种新的横截面方法在短爆炸窗口内分离共同和异质成分，蒙特卡洛模拟显示该方法对非单调功效问题稳健。
The left-tailed unit-root tests of the panel analysis of nonstationarity in idiosyncratic and common components (PANIC) proposed by Bai and Ng (2004, Econometrica 72, 1127–1177) have standard local asymptotic power. We assess the size and power properties of the right-tailed version of the PANIC tests when the common and/or the idiosyncratic components are moderately explosive. We find that, when an idiosyncratic component is moderately explosive, the tests for the common components may have considerable size distortions, and those for an idiosyncratic component may suffer from the nonmonotonic power problem. We provide an analytic explanation under the moderately local to unity framework developed by Phillips and Magdalinos (2007, Journal of Econometrics 136, 115–130). We then propose a new cross-sectional (CS) approach to disentangle the common and idiosyncratic components in a relatively short explosive window. Our Monte Carlo simulations show that the CS approach is robust to the nonmonotonic power problem.