利用蒙特卡洛分析方法,对目前文献中八种不同跳跃检验方法的检验水平和检验功效进行综合比较分析,并特别关注跳跃类型、市场微观结构噪声、零日内收益、日内周期性波动模式等对各种检验方法的影响。同时,为了克服直接利用非参数统计量进行检验时,因为多重检验而高估跳跃发生的次数的难题,将FDR阈值理论扩展到全部跳跃检验八种方法中,采用FDR阈值理论对价格跳跃检验中错误检验问题进行研究,发现FDR阈值方法能一定程度上减少错误检验问题,将错误检验率控制在FDR值内。
This paper performs a comprehensive Monte Carlo simulation comparison be- tween eight tests available in the literature to detect jumps in financial assets. The relative performance of eight tests is examined in a Monte Carlo simulation covering scenarios of both finite and infinite activity jumps, and stochastic volatility models with continuous and discon- tinuous volatility sample paths, this paper evaluates size and power properties of the proce- dures under alternative sampling frequencies, levels of volatility, persistence in volatility, degree of contamination with mierostructure noise, jump size and intensity. Whatever the jump detection test and the sampling frequency, a highly relevant number of spurious detec- tions remain because of multiple testing issues.