A0306
Title: Periodogram regression a two stage mixed effects approach for tropical cyclone frequency
Authors: Sourav Das - Curtin University (Australia) [presenting]
Guoqi Qian - The University of Melbourne (Australia)
Lyuyuan Zhang - University of Melbourne (Australia)
Abstract: Tropical cyclones (TC) are significant indicators of evolving climate dynamics. Two primary responses of interest are the cyclone frequency and intensity. A novel integrated modelling framework is proposed for the simultaneous modelling of TC frequency across several meteorological regions within Australasia. The key methodological insight is to model the second-order properties of multiple integer-valued time series in the frequency domain instead of parametric time domain models. A two-stage semiparametric approach is taken where large-scale environmental variation is modelled using generalized linear models while the stochastic variation, including spatial heterogeneity, is estimated using spectral analysis of time series under a hierarchical generating model. Using longitudinal data analysis, periodicities are jointly modeled in TC frequencies and their correlation with El Nino Southern Oscillation (ENSO) cycles, but also the spatial variation between regions. The fitted model is projected to obtain one-step-ahead forecasts using the principles of the best linear unbiased estimators. This semi-parametric approach avoids the uniqueness issues of parametric integer-valued time series modelling. Additional methodological advantages include tests for spatial heterogeneity and temporal second-order stationarity. The data analysis corroborates previous findings on the declining trend of tropical cyclone frequencies in the short term.
