Speaker |
Yanbo Tang |
|---|---|
Affiliation |
Imperial College London |
Date |
Wednesday, 03 December 2025 |
Time |
12:30-13:30 |
Location |
Ground floor lecture theatre, Sainsbury Wellcome Center, 25 Howland St, W1T 4JG |
Link |
https://ucl.zoom.us/j/99748820264 |
Event series |
Jump Trading/ELLIS CSML Seminar Series |
Abstract |
The use of integration methods is examined for approximating key quantities associated with complex posterior distributions and marginal likelihoods, together with their corresponding statistical guarantees. Our analysis differs from the usual approach, as we consider the integration of random functions rather than a fixed function, which is the typical focus in most numerical analysis works. We pay particular attention to two classes of integration techniques: quadrature methods and grid-based Monte Carlo methods. Their computational complexities are analyzed in high-dimensional settings. Finally, we provide illustrative examples that demonstrate the practical implementation of these methods. |
Biography |
I am a lecturer (assistant professor) in the statistics section at Imperial College London. I obtained my PhD from the University of Toronto in 2022 under the joint supervision of Nancy Reid and Daniel Roy. My main interests lie in statistical inference in the presence of a high number of nuisance parameters, and computational statistics. I am also interested in inference problems in statistical genetics and astrostatistics. |