통계연구소에서는 다음과 같이 통계 세미나를 개최하오니 많은 참여 바랍니다.
일시: 2011년 5월 25일(수) 오후 5시
장소: 고려대학교 정경관 201호
연사: Kai Wang Ng
(Head of the Department of Statistics and Actuarial Science,
The Univeristy of Hong Kong)
“The Converse of Bayes Theorem and some implications”
A Colloquium talk by Professor Kai Wang Ng
Bayesian inference has become more and more popular in all areas of research in the last 30 years or so. The related paradigm of inference is based on what is now called Bayes’ Theorem, or Bayes’ Formula, and was presented in a self-contained paper, “An Essay towards Solving a Problem in the Doctrine of Chances.” The author Rev. Thomas Bayes did not see it published in his lifetime and died on 17 April 1761. The manuscript was found among inherited papers by his friend, Richard Price, who communicated it to the Royal Society. The paper was read on 23 December 1763 and published in 49 pages by the Royal Society. It is still a mystery as why Bayes, himself a Fellow of the Royal Society since 1742, did not communicate the paper to the Royal Society in his lifetime. There have been interesting conjectures, including those offered by Sir R.A. Fisher in his writings and by S. M. Stigler in his 1983 article in American Statistician. In this colloquium the speaker presents his view that Bayes withheld his Essay because he saw the Converse of his own Theorem in the Essay and thus wanted more time to re-write it in those days of writing with feather and ink. The speaker shall explain his logic of the conjecture and demonstrate that the Converse can be derived with mathematics of Bayes’ time. Some other interesting implications of the Converse will also be presented.