Reports from the Heart of the Machine
Edit: Egg sauce
His "COX regression model" has profoundly influenced statistical research.
Last night, david Cox, a well-known British statistician, died at the age of 97.
David Cox is widely known for proposing the "COX regression model" and has profoundly influenced research in the field of statistics. Many spontaneously expressed grief and condolences on social media platforms:
Life of David Cox
Born in Birmingham, England in 1924, David Cox studied mathematics at St John's College, Cambridge, and received his PhD from the University of Leeds in 1949 under the direction of Henry Daniels and Bernard Welch.
From 1950 to 1956, David Cox worked in the Statistics Laboratory at the University of Cambridge. From 1956 to 1966 he was a "Reader" and professor of statistics at Birkbeck College, University of London. In 1966, he became Chair of the Department of Statistics at Imperial College London and later Chair of the Department of Mathematics. In 1988, he became Dean of Nuffield College and a member of the Department of Statistics at Oxford University, before officially retiring in 1994.
David Cox has made pioneering contributions to statistics and applied probability, including the Cox process and the far-reaching and widely used Cox proportional risk model.
David Cox was president of the International Statistical Association, the Bernoulli Society for Mathematical Statistics and Probability, and the Royal Statistical Society. He is also a fellow of the Royal Society and the British Academy of Social Sciences, a foreign member of the American Academy of Sciences and the Royal Danish Academy of Sciences.
For his significant contributions, David Cox was awarded the Guy Medal (1961) and the Gold Medal (1973) by the Royal Statistical Society, and was knighted by Queen Elizabeth II in 1985. In 2010, he was awarded the Copley Medal of the Royal Society for "pioneering contributions to statistical theory and applications". He was also the first to win the International Prize in Statistics (2017).
Cox regression model
The statistical field of survival analysis involves the time interval before a particular event, such as a mechanical failure or a patient's death. The rate at which failures occur or patients die here is called the risk function.
In the Cox proportional risk regression model introduced in 1972, David Cox proposed a risk function, which is divided into two parts: time dependence and time independence.
Thesis link: https://rss.onlinelibrary.wiley.com/share/XB97VAHIGECJZEBBBTWZ?target=10.1111/j.2517-6161.1972.tb00899.x
This model is often used in medical research to analyze the effect of one or more predetermined variables on patient survival time. By separating time-dependent inputs from time-independent inputs, the analysis of medical data has been greatly simplified, and the Cox model has been widely used in medical research. According to Google Scholar incomplete statistics, this article is currently cited more than 56612 times, and it is also the most used multi-factor analysis method in survival analysis to date.
In October 2014, cox regression became one of the "three most cited statistical papers" among the 100 most cited papers in Nature magazine.
In addition, David Cox is the author of numerous books in the field of statistics, including the theory of stochastic processes (co-authored with H.D. Miller, 1965), theoretical statistics (co-authored with D.v. Hinkley, 1974), survival data analysis (co-authored with David Oakes, 1984), and the principles of inferential statistics (2006).