Page 624 - Probability and Statistical Inference

14. Appendix 601

The Bayesians argued that Fisher’s theory of fiducial probability was not valid
without fully subscribing to the Bayesian principles. But, Fisher vehemently
defended his position that he never relied upon Bayesian arguments to develop
the concept of fiducial probability. He never caved in.
Fisher had close ties with W. S. Gosset who is perhaps better known
under the pseudonym “Student” than under his own name. Gosset graduated
(1899) with a first class degree in chemistry from New College in Oxford,
then joined the famous Guinness Brewery in Dublin as a brewer, and stayed
with this brewing firm for all his life, ultimately becoming the Head Brewer in
a new installation operated by the Guinness family at Park Royal, London in
1935. Gosset needed and developed statistical methods for small sample sizes
which he would then apply immediately to ascertain relationships between the
key ingredients in beer. Gosset’s path-breaking 1908 paper gave the founda-
tion of the t-distribution where he derived the probable error of a correlation
coefficient in 1908 and made several forceful conjectures, most of these be-
ing proven true later by Fisher (1915). For both Gosset and Fisher, making
scientific inferences and using small-sample statistics for experimental data
went hand-in-hand on a daily basis. The theory of statistics was essential for
applying statistics. There was no line drawn between the two. Fisher created
the foundation of this discipline. Rao (1992a) called Fisher “The founder of
modern statistics.” Fisher regarded Gosset very highly and he once described
Gosset as “Faraday of statistics.” Box (1987) pointed out various connec-
tions and collaborations among Guinness, Gosset and Fisher.
Fisher’s first book, Statistical Methods for Research Workers, was out in
1925. The Genetic Theory of Natural Selection, first appeared in 1930. The
Design of Experiments, appeared in 1935. The Statistical Tables for Biologi-
cal, Agricultural and Medical Research of R. A. Fisher and F. Yates was out
in 1938. Fisher’s three other books, The Theory of Inbreeding, Contributions
to Mathematical Statistics, followed by Statistical Methods and Scientific
Inference, respectively appeared in 1949, 1950 and 1956. These books have
served as landmarks in the history of the statistical science.
Fisher received many honors. The list includes, Fellow of the Royal Soci-
ety (1929), Honorary Fellow of the Indian Statistical Institute (1937), Royal
Medal (1938), Darwin Medal (1948), and Copley Medal (1955) of the Royal
Society, Guy Medal in Gold of the Royal Statistical Society (1946). He be-
came a Foreign Associate of the U.S. National Academy of Sciences (1948).
He was Knighted in 1952.
An extra-ordinary body of now well-known developments in modern
statistical theory and methods originated in the work of Fisher. One encoun-