| Speaker: | Guangming Pan, Nanyang Technological University |
| Host: | Qinwen Wang, School of Data Science, Fudan University |
| Time: | 17:00-18:00, December 12, 2019 |
| Location: | Zibin S301, Fudan University |
| Abstract: | This talk is about the spiked eigenvalues for high dimensional data with both cross-sectional dependence and dependent sample structure. We illustration its applications by investing whether we should implement PCA (factor model) for non-stationary data. A statistic is proposed for distinguishing between unit root models and non-stationary factor model. |
| Bio: | Guangming Pan is a Full Professor in the School of Physical & Mathematical Sciences at Nanyang Technological University. He got his PhD degree in mathematical statistics at University of Science and Technology of China. His main research interests include: random matrix theory, high dimensional statistics inference and applications of probability. |