Lightnews — Scholar-powered news

American Statistical Association History of Statistics @hos-asa.bsky.social · 17h

#OTD 1924 John A Nelder b (d 7 Aug 2010) FRS 1976, Guy Silver 1977 & Gold Medal 2005. Made major contributions to design of experiments. Head of Statistics at Rothamsted 1968-84. Awarded the inaugural ISI Karl Pearson Prize with Peter McCullagh for developing generalized linear models.

2

American Statistical Association History of Statistics @hos-asa.bsky.social · 2d

#OTD Faith Moors Williams (Lorimer) b (d 20 Sep 1958). ASA Fellow 1946. Director of the Office of Foreign Labor Conditions in the Bureau of Labor Statistics & one of the architects of the BLS cost-of-living index, later the US Consumer Price Index.

3

American Statistical Association History of Statistics @hos-asa.bsky.social · 3d

The phrase “Lies, damned lies, and statistics” is often attributed to Mark Twain who used it a lot, but he claimed to be quoting Benjamin Disraeli. However the phrase (and several variants) was already well-known: Carroll D. Wright (5th ASA president) was one of many users

American Statistical Association History of Statistics @hos-asa.bsky.social · 4d

#OTD 1841 Thomas Corwin Mendenhall b (d 22 Mar 1924) Father of “author profiling”, using word frequency distributions to identify individual writer styles. He kicked off the entire "Shakespeare was actually written by Francis Bacon or Kit Marlow" controversy although his data did not support it.

2

American Statistical Association History of Statistics @hos-asa.bsky.social · 5d

Counting-out rhymes used to select an item or person from a group (Eeny, meeny, miny, moe; Tinker, tailor; Eetam, peetam, penny pie) may have originated from the casting of lots for divination. They are not really 'random' selection processes as the results are determined by the starting selection

American Statistical Association History of Statistics @hos-asa.bsky.social · 6d

Today is the 75th anniversary of Alan’s Turing’s paper "Computing Machinery and Intelligence" published in journal Mind where he proposed the question, "Can machines think?" – the Turing Test

1 16 55

American Statistical Association History of Statistics @hos-asa.bsky.social · 7d

#OTD 1912 Dame Kathleen Ollerenshaw b (d 10 Aug 2014) 1st F president of Manchester Statistical Society 1981-3 (& member for 70 years, 1944-2014). An accomplished mathematician she used statistics to affect government social policies especially for improved school standards & girls’ education.

3

American Statistical Association History of Statistics @hos-asa.bsky.social · 8d

Math or Maths? Unlike the USA, in the UK the S is kept in the short version of “mathematics”. Adding an S to field of study nouns (physics, economics) began in the 18th c. These are mass nouns (entities that cannot be counted), so the S is not a pluraliser but a “deadbeat second S” (as in “mass”)

American Statistical Association History of Statistics @hos-asa.bsky.social · 9d

2/ Best known for eponymous Hotelling’s law lemma, T2 statistic & developer of principal component analysis. He was first to recognise the revolutionary importance of Fisher’s 'Statistical Methods for Research Workers' (which had received almost all hostile reviews when it first came out)

1

American Statistical Association History of Statistics @hos-asa.bsky.social · 9d

OTD 1895 Harold Hotelling b (d 26 Dec 1973) ASA Fellow 1937. Pioneer in mathematical statistics (multivariate analysis, hypothesis testing, correlation) & economics (competition, game-theory, depreciation, resource exhaustion), 1/2

1 1

American Statistical Association History of Statistics @hos-asa.bsky.social · 10d

#OTD 1937 Olimpiy Kvitkin (b 30 Oct 1874) chief Soviet statistician was shot on Stalin's orders. Labelled an "enemy of the people" for producing politically inconvenient data, his census data contradicted Stalin's official state numbers & showed the effect of the 1932-3 famine (Holodomor).

2 1

American Statistical Association History of Statistics @hos-asa.bsky.social · 11d

#OTD 1822 Jean-François Champollion announces the deciphering of Egyptian hieroglyphics using the Rosetta Stone. Although not a statistician, he used numerical pattern recognition to match frequency of symbols within cartouches to known Greek names, now standard codebreaking practice.

The Rosetta Stone can be seen in the British Museum London

4

American Statistical Association History of Statistics @hos-asa.bsky.social · 12d

#OTD 1876 Edith Abbott b (d 28 Jul 1957) ASA Fellow 1945. One of the 1st F deans in the US, she pioneered statistical methods applied to policy development for welfare, education & protections for immigrants, working women & children

1 2

American Statistical Association History of Statistics @hos-asa.bsky.social · 13d

#OTD 1893 Harold Cramér b (d 5 Oct 1985; Sweden) "one of the giants of statistical theory" ASA Fellow 1950, RSS Guy Gold Medal 1972. Best known for Cramér-Rao inequality, Cramér-von Mises statistics, Cramér-Levey theorem

3

American Statistical Association History of Statistics @hos-asa.bsky.social · 14d

2/ Metropolis came up with the cool name because Ulam used to talk about his uncle who liked to go to Monte Carlo to gamble

1 2

American Statistical Association History of Statistics @hos-asa.bsky.social · 14d

In Sept 1949 Stanislaw Ulam & Nicholas Metropolis introduced the method of Monte Carlo computer-intensive probability-based sampling. Ulam got the idea from playing solitaire during recovery from surgery. 1/2

1 3

American Statistical Association History of Statistics @hos-asa.bsky.social · 16d

3/ Representing the Physical Powers of Each Distinct Nation with Ease and Perspicuity. To which is added, a Similar Exhibition of the Ruling Powers of Hindoostan”. Can’t top that

2

American Statistical Association History of Statistics @hos-asa.bsky.social · 16d

2/ His magnum opus describing these methods was his 1801 book with the best-ever title “The Statistical Breviary; Shewing, on a Principle Entirely New, the Resources of Every State and Kingdom in Europe; Illustrated with Stained Copper-Plate Charts, ..

1 2

American Statistical Association History of Statistics @hos-asa.bsky.social · 16d

#OTD 1759 William Playfair b (d 11 Feb 1823) A Scottish engineer he invented or improved numerous statistical graphics methods still used today, including line, area, bar & pie charts. 1/3

1 2 4

American Statistical Association History of Statistics @hos-asa.bsky.social · Sep 1

AmStat History of Statistics @hos-asa.bsky.social is taking a break for a few weeks to attend the Royal Statistical Society conference & then go walkabout. Posts will resume the last week of September.

1

American Statistical Association History of Statistics @hos-asa.bsky.social · Aug 31

#OTD 31 Aug 1918 Major André-Louis Cholesky is killed in action. His matrix decomposition method for solving systems of n linear equations is published posthumously. Thirty years later in 1948 Alan Turing demonstrates the stability of the method for large-n machine computations.

American Statistical Association History of Statistics @hos-asa.bsky.social · Aug 31

#OTD 31 Aug 1907 Ruth Rice Puffer b (d 2 Sep 2002) ASA Fellow 1966, public health biostatistics and head of the Department of Health Statistics of the Pan American Health Organization . On their 100th anniversary PAHO listed her in the top 100 "public health heroes".

2

American Statistical Association History of Statistics @hos-asa.bsky.social · Aug 31

#OTD 1890 Julie E Backer b (d 31 Dec 1977) ISI Fellow 1948, bureau chief for the Norwegian Central Bureau of Statistics 1936–1956 & an influential researcher on public health, infant mortality, & population statistics

2

American Statistical Association History of Statistics @hos-asa.bsky.social · Aug 30

In 1605 Francis Bacon complains about researchers exaggerating scientific claims without substantive evidence & accepting claims of others as true without doing due diligence: he called this the “contract of error” (An Advancement of Learning)

1 4

American Statistical Association History of Statistics @hos-asa.bsky.social · Aug 29

In Aug 1934 the term ‘likelihood’ is first used in a Bayesian context by Harold Jeffreys as the "the theorem of Inverse Probability" where posterior probability is proportional to the product of the prior probability & the likelihood. Ronald Fisher had first used the term in its modern sense in 1921

1