Penny S. Reynolds, University of Florida College of Medicine
APRIL BIRTHDAYS
1667 “Physician in Ordinary” to Queen Anne, John Arbuthnot developed the first formal significance test (1710) to test the hypothesis of “the Exacte Balance” of human sex ratio at birth. His 1692 translation of Huygens’s De Ratiociniis in Ludo Aleae was the first work on probability in English.
1707 Leonhard Euler, called by Pierre-Simon de Laplace the “master of us all,” was prolific in every sense of the word (>860 publications, 13 children). He did pioneering work in graph theory, topology, astronomy, fluid mechanics, music theory, analytic geometry, trigonometry, calculus, and number theory. He created the whole form of modern mathematical thinking by inventing consistent notation and terminology; examples are summation Σ, Euler’s number e = 2.71828, i = √(-1), and the concept of functions f(x).
1777 Called “The Prince of Mathematicians,” Carl Friedrich Gauss was one of the most influential mathematicians of all time and is best known for work in probability, least squares, and numeric integration. He also made major contributions to theory of functions, normal distribution, number theory, differential geometry, theoretical astronomy, magnetism, and optics.
1787 Pierre Charles-Alexandre Louis, a pioneer in medical statistics, used methods that are the precursors of modern epidemiological practice: standardized patient histories; comparators; and grouping of cases by similarities in age, sex, diet, and disease severity and onset.
1878 Kirstine Smith was described by Karl Pearson as his most brilliant student. Invented the field of optimal designs. Her 1918 PhD thesis described G-optimal designs for polynomial regression.
1879 Edward Bidwell Wilson is best known for the Wilson score interval for binary variables, which anticipated the concept of confidence intervals, now attributed to Neyman and Pearson.
1891 Sir Harold Jeffreys (Guy Gold Medal, 1962) was a geophysicist best known for developing Bayesian theory of probability (Jeffrey’s prior). It received little attention for years due to the dominance of the Fisher-Neyman-Pearson school, which he opposed.
1894 Jerzy Neyman (ASA Fellow, 1942; Guy Gold Medal, 1966; ASA Wilks Award, 1968; US National Medal of Science, 1969) developed modern concepts of estimation and sampling. With Egon Pearson, he developed the hypothesis testing paradigm in 1933 and the modern confidence interval in 1937.
1895 Alexander Craig “Alec” Aitken is best known for his mathematical memory and fast mental calculations. With Harold Silverstone, he established the lower bound on the variance of an estimator, now known as the Cramér–Rao bound. Introduced generalized least squares and the now-standard vector-matrix notation for linear regression models in 1935.
1898 William O. Kermack is best known for the Kermack-McKendrick mathematical theory of epidemic disease spread, the origin of SIRS epidemiological models.
1903 Andrei Kolmogorov developed the nonparametric goodness of fit test in 1933, later extended by Smirnov. He was also a major developer of modern probability theory and laid the foundations of Markov random process theory.
1907 Henry Scheffé (ASA Fellow, 1951; ASA Vice President, 1954–1956; Fellow, International Statistical Institute; Fellow and President, Institute of Mathematical Statistics) Best known for the Scheffé method for multiple comparisons, ANOVA, and optimal tests for equality of variances. He also did important work on quality control, mixture designs, and nonparametric statistics.
1912 Herman Otto (H. O.) Hartley (ASA Fellow, 1953; ASA President, 1979; ENAR President, 1959; Wilks Medal, 1973) is best known for Hartley-Pearson tables and F-max test. He also made substantial contributions to sampling theory, optimization, variance component estimation, cancer drug safety, correspondence analysis, and sample survey design.
1916 Lee Cronbach (ASA Fellow, 1956) is best known for Cronbach’s α. He made contributions to measurement theory, test performance quantification, and concepts of validity.
1916 Claude Shannon was a Bell Labs engineer, father of communication and algorithmic information theory, and developer of the first experiments in AI. His work is fundamental to the invention of the internet, digital computers, smart phones, .zip files, and .jpg files.
1919 David Blackwell (ASA Fellow, 1962; ASA Vice President, 1978; Institute of Mathematical Statistics President, 1956; Royal Statistical Society Honorary Fellow, 1976; National Medal of Science (posthumous), 2014) was the first African American to be elected to ASA and National Academy of Sciences. Best known for Rao–Blackwell theorem and contributions to game theory, Bayesian statistics, and probability theory.
1923 Jacob Cohen (ASA Fellow, 1983) is best known for development of Cohen’s kappa, d, and h; his advocacy of power and effect size; and criticisms of misapplications of null hypothesis significance testing.
1927 Constance van Eeden (ASA Fellow, 1973; Institute of Mathematical Statistics Fellow, 1973; International Statistical Institute, 1978; Henri Willem Methorst Medal, 1999; Statistical Society of Canada Gold Medal, 1990; Statistical Society of Canada Honorary member, 2011) is best known for pioneering work in inference for nonparametrics, restricted parameter spaces, probability, and decision theory.
Events
1657 Christiaan Huygens publishes De Ratiociniis in Ludo (The Value of All Chances in Games of Fortune), the first printed work on the calculus of probabilities and the earliest to define and systematize concepts such as expectation. Some of the exercises he includes were later developed into probability distributions.
1742 Thomas Bayes is nominated to the Royal Society based on his “known merit, well skilled in geometry and all parts of mathematical and philosophical learning.” Best known for his eponymous theorem, though he never published it; his notes were edited posthumously by his friend Richard Price.
1755 Thomas Simpson reads his paper on error distributions, “On the Advantage of Taking the Mean of a Number of Observations in Practical Astronomy,” to the Royal Society. A milestone in statistical inference and the quantification of uncertainty, it is the earliest to show the arithmetic mean is nearer the ‘truth’ than any single observation (the law of large numbers).
1810 Pierre-Simon de Laplace reads his paper on the central limit theorem to the Académie de Sciences. He took almost 40 years to develop the method as a tool to solve other mathematical problems such as “the comet problem,” method of least squares, and risk in games of chance. However, he did not state an explicit theorem—that was precisely formulated around 1920 by others.
1835 Adolph Quetelet publishes his concept of ‘the average man,’ quantifies traits by the mean and the upper and lower limits of variation, and describes the Quetelet index. This index is later used as an indicator of health by Louis Dublin in 1912 and called the Body Mass Index in 1972.
1912 R. A. Fisher, a third-year undergraduate student at Cambridge, publishes his first mathematical paper, “On an Absolute Criterion for Fitting Frequency Curves.” This was an early version of what he would later call the method of maximum likelihood and features a critique of the method of moments favored by Pearson.
1922 Ronald Fisher publishes his paper, “On the Mathematical Foundations of Theoretical Statistics,” first read to the Royal Statistical Society on November 17, 1921. It proposes a unifying theory of statistics and marks the beginning of modern mathematical statistics. The focus is on problems of estimation and distribution and introduces the key ideas of consistency, efficiency, sufficiency, and information.
1942 Sir Maurice Kendall (1907–1983) reads his paper, “On the Future of Statistics,” before the Royal Statistical Society. He predicts the need for “new types” of statisticians, from general practitioners to pure mathematicians, and the rising importance of “statistical machines” (computers).
1959 Nathan Mantel and William Haenzel publish the Mantel-Haenszel test for analyzing stratified or matched categorical data obtained from retrospective observational studies. It is now one of the most-cited statistics papers of all time.
1964 George Box and David Cox read their paper “An Analysis of Transformations” at the Royal Statistical Society meeting. The paper began with their names, then found the topic. Everyone thought it would be hilarious to have them write a paper together: “Box and Cox” is a colloquialism in the UK for taking turns, based on an 1847 one-act comic play about two men tricked by their landlady into occupying the same room without being aware of each other’s existence.
1994 Douglas Altman and Martin Bland publish the first of British Medical Journal’s Statistics Notes in statistics education. The notes have been called the finest series of short articles on the use of statistics.
View the references for this month’s Time Series.
Leave a Reply