一类股市波动性预测模型的多变点检验

讨论了一类股市波动性预测模型的多变点检验问题.基于累积和(CUSUM)统计量,提出一种新的波动性预测模型多变点的拟似然比检验,在原假设下给出统计量的极限分布及渐近临界值的解析表达式.多变点的递归检验是本文的一个主要方面,即在检验时允许原假设含有变点,备择假设比原假设多一个变点,并在递归检验的过程中同时得到变点时刻与变点个数的一致估计,因此为处理含多变点的金融数据时确定变点个数提供了一种建模策略.数值模拟与实例分析说明了方法的有效性.

Multiple Change Point Test of a Volatility Forecasting Models in the Stock Market

This paper studied the problem of multiple change point test of a volatility forecasting models in the stock market.We propose a new pseudo likelihood ratio test which is based on the cumulative sum(CUSUM) statistic.The limiting distribution of the test under null hypothesis is present.In addition,We derive analytical expression for asymptotic critical value.The main aspects are the procedure of the recursive test which allows one to test the null hypothesis of changes versus the alternative of one more changes than the null.The consistent estimator of the date and the number of breaks is instantaneously obtained via the test procedure.Therefore a general modelling strategy to consistently determine the appropriate number of changes in financial data series is also obtained.A simulation study and real data analysis support the validity of our test.