对平稳过程趋势项中的变点检测问题,基于最小二乘估计残差构造CUSUM型统计量.同时,在原假设下得到了其渐近分布,在备择假设下证明了该检验的容许性.蒙特卡罗数值模拟表明,文中所给检验在检验趋势项系数变点时具有较好的势,且犯第一类错误的概率较小.
A CUSUM test for the detection of change in stochastic trend with stationary innovations is proposed.The asymptotic distribution is obtained under the null hypothesis,and the consistency of the proposed test is established.The Monte Carlo simulations demonstrate that the proposed test has a good size and high power.