2020/12/27 by Eiji Kurozumi, Kurozumi, Eiji, Anton Skrobotov +3
Economics, Econometrics and Finance · #Complex Systems and Time Series Analysis
paper · pdf · doi:10.48550/arxiv.2012.13937
This paper is devoted to testing for the explosive bubble under time-varying\nnon-stationary volatility. Because the limiting distribution of the seminal\nPhillips et al. (2011) test depends on the variance function and usually\nrequires a bootstrap implementation under heteroskedasticity, we construct the\ntest based on a deformation of the time domain. The proposed test is\nasymptotically pivotal under the null hypothesis and its limiting distribution\ncoincides with that of the standard test under homoskedasticity, so that the\ntest does not require computationally extensive methods for inference.\nAppealing finite sample properties are demonstrated through Monte-Carlo\nsimulations. An empirical application demonstrates that the upsurge behavior of\ncryptocurrency time series in the middle of the sample is partially explained\nby the volatility change.\n