重尾过程协整检验的Bootstrap逼近

Bootstrap approximation for cointegration tests in heavy- tailed observations

重尾过程的协整检验统计量渐近分布含有不可估计的重尾指数α,针对重尾过程的协整检验运用Bootstrap抽样算法,在不估计重尾指数α的情况下,计算了该检验统计量的临界值,并且证明了该算法在理论上的合理性.Monte Calo模拟说明该方法有效.

It is known that the asymptotic distribution of a cointegration test statistic in heavy - tailed observations has the inestimable index parameter. This research uses the Bootstrap method in the cointegration test in heavy - tailed processes, calculates the critical value of the cointegration test without estimating the index parameter , and proves the theoretical validity of this Bootstrap method. MonteCalo simulations demonstrate the efficiency of the proposed method.