the Chambers Medal of the Royal Statistical Society for his contributions and work for the
society.
Wishart, J ohn ( 1 898^1956): Born in Montrose, Wishart studied mathematics and natural philos-
ophy at Edinburgh University obtaining a first class honours degree in 1922. He began his
career as a mathematics master at the West Leeds High School (1922–1924) and then joined
University College, London as Research Assistant to
Karl Pearson
. In 1928 he joined
Fisher
at Rothamsted and worked on a number of topics crucial to the development of multivariate
analysis, for example, the derivation of the generalized product moment distribution and the
properties of the distribution of the
multiple correlation coefficient
. Wishart obtained a
readership at Cambridge in 1931 and worked on building up the Cambridge Statistical
Laboratory. He died on 14 July 1956 in Acapulco, Mexico.
Within groups matrix of sums of squares and cross-products: See multivariate
analysis of variance.
Within groups mean square: See mean squares.
Within groups sum of squares: See analysis of variance.
W˛hler curve: Synonymous with S–N curve.
Wold, Herman Ole Andreas (1908^1992): Born at Skien, Norway on Christmas Day 1908.
Wold’s family moved to Sweden in 1912 and he began his studies at Stockholm University
in 1927. His doctoral thesis entitled ‘A study in the analysis of stationary time series’ was
completed in 1938, and contained what became known as Wold’s decomposition, the
decomposition of a time series into the sum of a purely non-deterministic component and
a deterministic component. In 1942 Wold became Professor of Statistics at the University of
Uppsala where he remained until 1970. During this time he developed a mechanism which
explained the
Pareto distribution
of wealth and, somewhat disguised, introduced the concept
of latent variable models. Wold died on 16 February 1992.
Wold’s decomposition theorem: A theorem that states that any
stationary stochastic process
can
be decomposed into a deterministic part and a non-deterministic part, where the two components
are uncorrelated and the non-deterministic part can be represented by a moving average process.
[Time Series Analysis, 1994, J. D. Hamilton, Princeton University Press, Princeton, NJ.]
Wolfowitz, Jacob (1910^19 81 ): Born in Warsaw, Poland, Wolfowitz came to the United States
with his family in 1920. He graduated from the College of the City of New York in 1931 and
obtained a Ph.D. in mathematics from New York University in 1942. In 1945 Wolfowitz
became an associate professor at the University of North Carolina at Chapel Hill, and in 1951
joined the Department of Mathematics at Cornell University. Wolfowitz made many important
contributions in mathematical statistics, particularly in the area of asymptotics, but also
contributed to probability theory and coding theory. He died in Tampa, Florida on 16 July 1981.
Woolf’s estimator: An estimator of the common
odds ratio
in a series of
two-by-two contingency
tables
. Not often used because it cannot be calculated if any cell in any of the tables is zero.
[Biometrical Journal, 2007, 29, 369–374.]
Worcester, Jane (1910^1989): Worcester gained a first degree in 1931 and joined the
Department of Biostatistics at Harvard School of Public Health as a mathematical computing
assistant. She remained in the department until her retirement in 1977, becoming during this
time the first female Chair of Biostatistics. During her 46 years in the department she
devoted her time to research, teaching and consultancy. Worcester died on 8 October 1989
in Falmouth, Massachussetts.
462