Efficient Leverage Score Sampling for the Analysis of Big Time Series Data
Seminar/Forum
We use randomized numerical linear algebra techniques to develop a new fast algorithm to estimate the leverage scores of an autoregressive model in big data regimes. We show that the accuracy of approximations lies within $(1 + O{\varepsilon})$ of the true leverage scores with high probability. These theoretical results are exploited to fit an appropriate autoregressive model to big time series data and approximately estimate its parameters by sampling the data matrix based on estimated leverage scores. Empirical results on largescale synthetic as well as real data highly support the theoretical results and reveal the efficacy of this new approach.
Presenter

Dr Ali Eshragh, Senior Lecturer
Dr Ali Eshragh
Senior Lecturer
University of NewcastleAli Eshragh is a senior lecturer in Statistics and Optimisation at the University of Newcastle. His research is primarily focused on the areas of Applied Probability, Statistical Modelling and Optimisation. He has received several awards for his excellence in teaching and research. Recently, he was awarded the Australian Society for Operations Research Rising Star Award. Ali is a chair and organiser of the first ongoing yearly workshop titled “Data Science Down Under” with the inaugural theme of Randomised Numerical Linear Algebra, that will be held in Newcastle, 812 December 2019.