Title: Testing for jumps: An increment censoring approach in boundary-hitting intrinsic time
Authors: Yifan Li - The University of Manchester (United Kingdom)
Ingmar Nolte - Lancaster University (United Kingdom)
Sandra Nolte - Lancaster University (United Kingdom)
Shifan Yu - Lancaster University (United Kingdom) [presenting]
Abstract: A novel nonparametric test is proposed to determine whether finite-activity jumps are present in a discretely observed price process or not. We use the concept of censored realized variation for the observations sampled at hitting times with respect to a symmetric double barrier to construct our test statistics for a univariate Ito semimartingale. The test statistics diverge to infinity if jumps are present and have a normal distribution otherwise. We use the Monte Carlo simulations to investigate the performance of our new tests in a range of empirically relevant scenarios and comparing them with other nonparametric jump tests based on calendar time sampling. We finally present the empirical results using the real-world high-frequency financial data.