FBBVA-16-ciencias-Cox-1

David Cox

FRONTIERS OF KNOWLEDGE LAUREATE

Basic Sciences

9th edition

The BBVA Foundation Frontiers of Knowledge Award in the Basic Sciences category goes, in this ninth edition, to mathematicians David Cox and Bradley Efron, for their pioneering and hugely influential contributions to the theory and practice of statistical science through the invention of powerful techniques in pervasive use, including the Cox regression and Efron’s bootstrap method.

CITATION (EXCERPT)

The BBVA Foundation Frontiers of Knowledge Award in Basic Sciences goes, in this ninth edition, to David Cox and Bradley Efron for their pioneering and hugely influential contributions to the theory and practice of statistics. Their methods are used on a daily basis in the practice of statistical science, and have made an enormous impact in all the sciences which rely on the analysis of data.

These techniques have found pervasive use. Examples include conclusions that even a year of smoking cessation decreases mortality, the development of medical measurement instruments, and analysis of clinical, economic and astrophysics data.

David Cox invented the proportional hazards model for the analysis of survival data. This “Cox model” is a fundamental tool for modeling lifetime distributions, and is widely used in many areas, including cancer research, epidemiology, medicine, economics, sociology and industrial testing. The Cox model elegantly accommodates observations that are “censored” – for example, when the study ends, or subjects are lost to follow-up – and allows for the estimation of treatment effects while adjusting for other variables that may affect the outcome.

Bradley Efron invented the “bootstrap” method, a computer-intensive resampling technique for approximating the distribution of statistical estimators extracted from data. The magic of this procedure is that it can be used with novel statistics, for which theoretical results are elusive. The bootstrap is widely used across all disciplines for estimating standard errors, error rates, confidence intervals, and hypothesis testing.