Arthur P. Dempster

From Wikipedia, the free encyclopedia
Jump to: navigation, search
Prof Dempster at the workshop on Belief Function Theory (Brest 1 April 2010).

Arthur Pentland Dempster (born 1929) is a Professor Emeritus in the Harvard University Department of Statistics. He was one of four faculty when the department was founded in 1957.[1]

He was a Putnam Fellow in 1951. He obtained his Ph.D. from Princeton University in 1956. His thesis, titled The two-sample multivariate problem in the degenerate case, was written under the supervision of John Tukey.

Among his contributions to statistics are the Dempster–Shafer theory and the EM algorithm.


  • A. P. Dempster, N. M. Laird and D. B. Rubin. (1977). "Maximum Likelihood from Incomplete Data via the EM Algorithm," Journal of the Royal Statistical Society, B, vol. 39, no. 1, pp. 1–38. (The authors collected a variety of maximum likelihood problems and methods of solving these problems that occurred in the literature. They found that all of these methods had some ideas in common and they named it the EM Algorithm, standing for "Expectation, Maximization".)


  1. ^ Date information sourced from Library of Congress Authorities data, via corresponding WorldCat Identities linked authority file (LAF) .

External links[edit]