Charles stone

from Wikipedia, the free encyclopedia

Charles M. Stein (born March 22, 1920 in Brooklyn , New York , † November 24, 2016 in Fremont , California ) was an American statistician .

Life

Stein graduated from Columbia University . During World War II worked in weather forecasting for the US Air Force and came into contact with statistical work. After the war he received his doctorate in 1947 from Columbia University under Abraham Wald . He was a professor at Stanford University .

Stein had been a member of the National Academy of Sciences since 1975 . In 1966 he gave a plenary lecture at the International Congress of Mathematicians (ICM) in Moscow (Some recent developments in statistics). He was Wald and Neyman Lecturer at the US Institute of Mathematical Statistics .

He developed the Stein method for approximating probability distributions (with limits for the errors). It emerged in the late 1960s when he was looking for a new way in his lectures to prove the “combinatorial” central limit theorem of Wald and Wolfowitz. An analogous method for the Poisson approximation was developed by Stein's student Louis HY Chen and is now known as the Stein-Chen method .

He wrote a paradox in statistical decision theory (1955) that if three or more parameters are to be estimated, methods with a combined estimation of the parameters are more precise than the usual methods that estimate the parameters separately. The paradox initially met with opposition and sparked heated debates.

He also showed Stein's lemma , which provides an estimate of the speed of convergence of power in Neyman-Pearson tests with increasing sample size. In addition, Barankin's and Stein's theorem on the structure of locally minimal estimators is named after him.

Fonts

  • Approximate computation of expectations, Hayward, California, Institute of Mathematical Statistics Lecturesnotes, 1986

Web links

See also

Individual evidence

  1. ^ Obituary in Stanford News , accessed January 30, 2017
  2. Stein: A bound for the error in the normal approximation to the distribution of a sum of dependent random variables, "Proceedings of the Sixth Berkeley Symposium on Mathematical Statistics and Probability 1972. See, for example, Barbour, Chen (editor) An introduction to Stein ´s Method , World Scientific, 2005
  3. Stein "Inadmissibility of the usual estimator for the mean of a multivariate distribution , Proceedings of the Third Berkeley Symposium on Mathematical Statistics and Probability, Vol. 1, 1956, pp. 197-206. See also Efron, Morris Stein's Paradox in Statistics , Scientific American, Vol. 236, 1977, Issue 5, online here PDF file