| Dr Heather Battey (Imperial College London) |
| Fri 03 May 2019, 15:05 - 16:00 |
| JCMB 5323 |
If you have a question about this talk, please contact: Tim Cannings (tcannin2)
The lasso and its variants are powerful methods for regression analysis when there are a small number of study individuals and a large number of potential explanatory variables, a small number of which having a real effect. There results a single model, while there may be several models equally compatible with the data. I will outline a different approach whose aim is essentially a confidence set of models. A heuristic probabilistic explanation for the success of the procedure will be given. The talk is based on joint work with David R Cox:
Cox, D.R. and Battey, H.S. (2017) Large numbers of explanatory variables, a semi-descriptive analysis. Proc. Nat. Acad. Sci., 114 (32), 8592-8595.
Battey, H.S. and Cox, D.R. (2018) Large numbers of explanatory variables: a probabilistic assessment. Proc. R. Soc. Lond. A., 474