Title: A bootstrap-based KPSS test for functional time series
Authors: Yichao Chen - Nanyang Technological University (Singapore) [presenting]
Chi Seng Pun - Nanyang Technological University (Singapore)
Abstract: The bootstrap method is applied to the KPSS test of functional time series to estimate the limit distribution of the test statistic when the unobserved noises of original sample are independent. We find that bootstrap method makes the testing process faster and more efficient than the methods have found, especially when sample size $N$ are not very big (no more than 70). The convergence of the bootstrap test statistic is established in a general probability space and then use simulation study to present the efficiency of our methods in KPSS test of functional time series.