Leonard J. Savage

Leonard Jimmie Savage , called Jimmie Savage, but he published as Leonard Savage, (originally Leonard Ogashevitz, born November 20, 1917 in Detroit , † November 1, 1971 in New Haven , Connecticut ) was an American statistician and mathematician.

Life

The name change from Ogashevitz to Savage was first carried out by his father in 1920, Leonard J. Savage took over when he was researching on secret government projects. Savage studied chemical engineering at Wayne University in Detroit and then at the University of Michigan . Since he started a fire in the chemistry laboratory because of his extreme ametropia, he was unable to continue studying chemistry and switched to physics and finally to mathematics. In 1938 he made his bachelor's degree as a student of Raymond Wilder . In 1941 he received his doctorate from Sumner Myers ( The Application of Vectorial Methods to the Study of Distance Spaces ). He then worked at the Institute for Advanced Study (IAS) in 1941/42 . In 1942/43 he was an instructor at Cornell University and then undertook military research tasks at Brown University and the statistical research group at Columbia University during the Second World War . During his time at the IAS he was a close associate of John von Neumann .

In 1944 he began to work with statistics. In 1945 he spent a year with Richard Courant at New York University and then on a Rockefeller scholarship at the University of Chicago, where he worked with Milton Friedman (whom he already met at Columbia University) in economics and with Paul Halmos . The Friedman-Savage utility function is named after a study by Friedman and Savage. In 1949 he was one of the founders of the Statistics Faculty in Chicago with Allen Wallis . In 1951/52 he was a Guggenheim Fellow and Fulbright Fellow in Paris and Cambridge .

In 1954 he became a professor in Chicago and was from 1956 to 1959 head of the faculty of statistics. In 1959 he gave lectures in London that led to a seminar (in which Egon Pearson took part), which was also published as a book. In 1960 he became a professor at the University of Michigan and in 1964 at Yale University . There he worked closely with Frank Anscombe (1918-2001).

Among other things, he dealt with the basics of statistics (also from a philosophical point of view), Bayesian statistics and statistics in game theory as well as applications in economics. He used the Bayesian approach to statistics for his foundations of statistics, that is, the subjective view of probabilities as expectations of a player, for example. He was influenced by Bruno de Finetti .

1957/58 he was President of the Institute of Mathematical Statistics. In 1963 he received an honorary doctorate from the University of Rochester. In 1958 he was invited speaker at the International Congress of Mathematicians in Edinburgh ( Recent tendencies in the foundations of statistics ).

He was married twice, his first marriage from 1938 until his divorce in 1964. From this first marriage he had two sons. His son, Samuel Linton Savage, who received his doctorate in computer science from Yale University in 1973 , wrote the book The flaw of averages . His brother Richard Savage (* 1925) was also a well-known statistician.

Fonts

• The foundations of Statistics , Wiley 1954
• with Lester Dubins How to gamble if you must: Inequalities for stochastic processes , McGraw Hill 1965

See also

• Halmos-Savage theorem

Web links

• John J. O'Connor, Edmund F. Robertson :  Leonard J. Savage. In: MacTutor History of Mathematics archive .
• Tales of Statisticians: Jimmie Savage, by E. Bruce Brooks

Individual evidence

1. ^ Mathematics Genealogy Project
2. ^ Friedman, Savage Utility Analysis of Choices Involving Risk , Journal of Political Economy, Volume 56, 1948, pp. 279-304
3. ^ Foundations of statistical inference; a discussion opened by LJ Savage at a meeting of the Joint Statistics Seminar, Birkbeck and Imperial Colleges, in the University of London , London, Methuen, New York, Wiley 1962