Robust Rank Tests of the Unit Root Hypothesis
提出基于回归秩得分过程的秩检验方法，用于经济时间序列的单位根检验。相比最小二乘法，该检验在零假设下渐近正态，且对厚尾创新过程具有更优的检验功效。
We consider a family of rank tests based on the regression rank score process introduced by Gutenbrunner and Jureckova' (1992) to test the unit root hypothesis in economic time series. In contrast to tests based on least squares methods, the rank tests are asymptotically Gaussian under the null hypothesis, and have excellent power-particularly under innovation processes exhibiting heavy tails.