Jerzy Neyman

April 16, 1894-August 5, 1981


Jerzy Neyman is considered to be one of great founders of modern statistics. He made large contributions in probability theory, testing hypothesis, confidence intervals, generalized chi-square, and other areas of mathematical statistics. He was enthu siastic about his work because he wanted to " find outí and study " how to find out what we need to know." His work would make an impact on fields ranging from astronomy and agriculture through biology and weather to social insurance. < /P>

Born Jerzy Splawa-Neyman(later dropping Splawa at age 30) in Bendery, Russia, Neyman had an early education by a governesses in French and German. This made Neyman proficient in many languages. In 1912, he entered the University of Kharkov to stu dy physics and mathematics. While studying at Kharkov, Neyman was taught by S.N. Bernstein. Bernstein introduced him to Karl Pearsonís Grammar of Science. Later Neyman would say that this influenced his development, but this was not the main i nterest of his during his studies. He was really interested in the research in measure theory of Lebesgue. This was the subject of most of his early papers.

In 1921, Neyman was forced to move to Poland to due to the war between Poland and Russia. Neyman was 27 at the time. In Poland, Neyman worked with W. Sierpinski before moving to London in 1924. Neyman studied under Karl Pearson while in London. He also made contacts with Egon Pearson, R. A. Fisher, and W. S. Gosset while at University College. By 1934, Karl Pearson had retired and his department was divided between his son Egon and Fisher. Egon invited Neyman to work with him. They worked on the theory of testing hypotheses. They supplied logical foundation and mathematical rigor to the theory that was missing in previous methodology. Their work was disputed by some mathematicians, including Fisher. The Neyman-Pearson ideas eventually spr ead throughout mathematics. Their ideas made sure that samples were large enough to avoid false representation.

Neyman also developed a theory of survey sampling in 1934. He used a theoretical basis for using probability sampling for cluster samples with a method for estimating the accompanying variances for the clustered samples. His results were used to a sampling survey of Polish labor . Neyman went to lecture about his theory in the United States.

The theory of estimation by confidence sets was Neymanís next topic of choice. He used confidence intervals to guarantee that the probability of covering the true value of the parameter to be estimated was at least equal to a preassigned value cal led the confidence coefficient. His uses soon appeared in many textbooks and works on statistical methodology.

In 1937 Neyman accepted a position at the University of California-Berkeley. He was asked to start a statistics department at Berkeley. Many people question his decision, but Neyman took the position because he was fearful of Hitler and the start of World War II. Is was at Berkeley, were he spent half of his life, that he came up with his BAN(best asymptotically normal) estimation theory. The BAN estimates now widely used in a manner similar to the use of least squares.

Neyman made a large impact in statistics throughout his lifetime. He received numerous awards and degrees for his work. He a mathematician who wanted to ensure that science was not obscured by political expedience or commercialism.




Encyclopedia of Statistical Science, Vol. 6, John Wiley & Sons, 1985. Pp. 215-221