Seasonality (or periodicity) and trend are features describing an observed sequence, and extracting these features is an important issue in many scientific fields. However, it is not an easy task for existing methods to analyze simultaneously the trend and dynamics of the seasonality such as time-varying frequency and amplitude, since the adaptivity of the analysis to such dynamics and robustness to heteroscedastic, dependent errors is not guaranteed. These tasks become even more challenging when there exist multiple seasonal components. Chen et al. (2013) propose a nonparametric model to describe the dynamics of multi-component seasonality, and investigate the Synchrosqueezing transform (SST) in extracting these features in the presence of a trend and heteroscedastic, dependent errors.
The identifiability problem of the model is studied and the model is compared with the methods EMD and TBATS, which were developed for similar problems. Also, results are provided of a series of simulations and the extraction of modes from a time series of the movement of japanese quails.
R, time series, time series decomposition, trend, seasonality, noise, wavelets, synchrosqueezing, EMD, TBATS, simulations
Student: Sofía Nieva
Director: Dr. Ana Georgina Flesia
Departament of Matemathics
Faculty of Exact and Natural Sciences
University of Buenos Aires