My specialization is in the area of time series analysis, particularly in multivariate and functional time series. My thesis research focuses on the inferences of functional time series, including prediction, bootstrap approximation and spectral density estimation.
Additionally, I have interned as a data scientist at Huawei R&D center in Santa Clara, CA, in summer 2016, where I investigated methods and designed UI of time series anomaly detection; See the Demos here, [Single and Multiple series outlier detection], [Online outliers detection].
- 'Kernel estimates of nonparametric functional autoregression models and their bootstrap approximation.'
T. Zhu and D. N. Politis, to appear in Electronic Journal of Statistics, 2017
- 'Higher-order accurate spectral density estimation of functional time series.'
T. Zhu and D. N. Politis, working paper, 2017
Talks and Posters
- 'Kernel estimation of nonparametric functional autoregression', [Poster].
Conference of Random Process and Time Sereis, in honor of Murray Rosenblatt, UCSD, Oct 2016.
- 'Kernel estimation of nonparametric functional autoregression and its bootstrap approximation', [Slides].
Candidacy talk, UCSD, June 2015.
- 'Time Series Outlier Detection', [Slides].
Huawei R&D center, Santa Clara, CA, July 2016.
- 'Online Outlier Detection for Time Series', [Slides].
Huawei R&D center, Santa Clara, CA, September 2016.
- 'Multilinear Principal Component Analysis of Tensor Objects', [Slides].
Master's final defense, Texas A&M, College Station, TX, July 2012.
- Electronic Journal of Statistics
- Journal of Time Series Analysis