Charity Commission charity number 313536
(Photograph by Denise Till)
The Fisher Memorial Trust was set up to promote interest in the life and work of the great
statistician, evolutionary biologist and geneticist, Sir Ronald Aylmer Fisher (1890-1962) and to
maintain his scientific legacy by encouraging discussion of the scientific fields in which he was
This talk will attempt to give the historical context of randomization and randomization-based inference from Fisher to the present day, including newer concepts such as response-adaptive, covariate-adaptive, and covariate-adjusted response-adaptive (CARA) randomization. It will be challenging to condense a year of material into one hour, but a devoted Fisherian should be able to be efficient and sufficient.
An Obituary of Fisher is available here Sankhya Obituary by kind permission of the Indian Statistical Insitute (ISI).
The 1990 Centenary Fisher Lecture delivered at the Indian Statistical Institute on 30 December 1990 by Partha Majumder is avalable here
by kind permission of Current Science and the Fisher Memorial Trust 1990 memorial Fisher lecture by Walter Bodmer
is available here Genetic sequences. An account
by Anthony Edwards of Fisher's eugenic theories can be found here The thoughtful eugenist 2020
There is discussion by Brian Charlesworth of the Eugenics Society 1933 view of eugenics in Nazi Germany here
R.A. Fisher, The Eugenics Society, and the Nazis in 1933
An article by the 40th Fisher Lecturer, Professor Kanti Mardia, relating to the subject of his lecture and entitled Fisher's Pioneering work on Discriminant Analysis and its Impact on AI is now available as a preprint.
|Prof Rosemary Bailey
|Dr Heather Battey***
|Mr Andrew Mead
|Prof Brian Charlesworth****, FRS
|Prof Ian Tomlinson, FRS**
|Sir Walter Bodmer, FRS
|Royal Statistical Society
|Prof Stephen Senn
|Royal Statistical Society
|Prof Christina Yap *
* Dr Andrew Sutherland until the end of 2020. Professor Christina Yap from 27 May 2022
** Replaced Professor Adam Eyre-Walker in 2021
*** Professor Vern Farewell until May 2021
**** Took over from Sir Walter Bodmer as chairman on November 18 2021
Biosketches of FMT members
|Dr. F. Yates
|Computers, the second revolution in statistics.
|Dr. R. R. Race
|Blood groups in human genetics
|Prof. E. A. Cornish
|Developments from the Fisher-Cornish expansions.
|Prof. K. Mather
|Prof. G. A. Barnard
|Statistical inference and its historical development
|Prof. L. L. Cavalli-Sforza
|Cultural versus biological evolution
|Prof. R. Hide
|Motions in planetary fluids
|Prof. D. J. Finney
|Bioassay and the practice of statistical inference
|Prof. J. Maynard Smith
|The evolution of the sex ratio
|Prof. J. H. Bennett
|R. A. Fisher and The Genetical Theory of Natural Selection
|Prof. S. Karlin
|Kin selection and altruism
|Prof. D. R. Cox
|Regression and the design of experiments
|Prof. S. M. Stigler
|Francis Galton and the unravelling of the normal world
|Prof. G. E. Box
|Quality improvement, an expanding domain for the application of scientific method
|Sir Walter Bodmer
|Prof. D. Lindley
|Statistics of the market place
|Prof. A. J. Jeffreys
|Molecular sleuthing: the story of genetic fingerprinting
|Dr. A. W. F. Edwards
|Fiducial inference and the fundamental theorem of natural selection
|Prof. M. J. R. Healy
|The life and work of Frank Yates
|Prof. J. A. Nelder
|Computers and statistics: the continuing revolution.
|Sir John Kingman
|Mathematics of genetic diversity: before and after DNA.
|Prof. B Efron
|The essential Fisher
|Sir Richard Doll
|Proof of causality: Deductions from epidemiological evidence
|Dr. Oliver Mayo
|The realisation of Fisher's research programme
|Prof. Warren Ewens
|Statistics and the transformation from genetics to genomics
|Prof. Adrian Smith
|Towards an evidence-based society: the role of statistical thinking
|1953: an unrecognized summit in human genetic linkage analysis
|Prof. R.A. Bailey
|Design of dose-escalation trials
| Prof B Charlesworth and
Prof D Charlesworth
|Fisher and modern evolutionary genetics
|Prof Philip Dawid
|Causal inference from experimental data
|Prof Peter Donnelly
|Genetic variation in human health and disease
|Prof David Spiegelhalter
|Putting life into numbers: the highs and lows of communicating statistics to the public
|Prof Bill Hill
|Applying quantitative genetic and genomic information to animal improvement
|Prof Peter McCullagh
|Empirical phenomena and universal laws
|Prof Nancy Reid
|Statistical science and data science: where do we go from here?
|Prof Stephen Senn
|And thereby hangs a tail: the strange history of P-values
|Prof Joe Felsenstein
|Is there a more fundamental theorem of natural selection?
|Prof Michael Goddard
|The genetic architecture of complex traits
|Prof Brian Cullis and Prof Alison Smith
|Prof Kanti Mardia
|Fisher's Legacy of Multivariate Analysis, Statistics on Manifolds, and Beyond
Professor Mardia gave an expert and often humorous account of Fisher's various contributions to spatial and multivariate analysis, covering not only the history of the subject but also modern developments.
One of the features of Fisher's work is that the starting point was often a motivating application from scientists and applied statisticians. For Directional Statistics, this was the problem of pole reversal raised by a geologist Mr J. Hospers. For Discriminant Analysis it was a set of cranial measurements taken by Mr E. R. Martin, who applied Discriminant Analysis to sex differences in measurements of the mandible. Further motivation was through the work of Miss Mildred Barnard in collecting and analysing a time series of skull measurements. However, in his 1936 paper, he first time introduces now celebrated iris data.
In this talk, first I will describe the Fisher distribution and reanalyse his geological data. He mentioned in the paper that it has two goals: the first to explore methodology for the analysis of widely dispersed measurements of direction such as frequently arise in geology; the second goal to illustrate the correct a pplication of fiducial inference. Regarding, the first goal the Fisher-von Mises distribution has become an essential tool for analysing spherical data but regarding his second goal on fiducial inference his achievement remains uncertain. Incidentally, Fisher derives, the von Mises distribution on the circle independently in 1959 as another example to illustrate the fiducial principle rather than analysing any circular data.
We will discuss the extension of the Fisher distribution on the sphere to hypersphere which is now known as the von Mises-Fisher distribution. In general, the maximum likelihood methods for directional distributions are not computationally straightforward and a new approach of the score matching estimate will be presented.
Coming back to the topic of discriminant analysis, historically, he wrote four papers on this topic during 1936-1940 and connected with the pioneering work in the same period of Hotelling and Mahalanobis. We revisit the famous iris data and answer one of the assumptions he points out in his work that he admits that he is assuming normality to assess the misclassification error, in particular. One of the most popular tests of normality is through my measures of multivariate skewness and kurtosis and we give evidence that his assumption was well founded. We also indicate how the subject has moved due to the computer revolution and there are now new methods such as kernel classifier, classification trees, support vector machine, neural networks to carry out discriminant analysis. His work is a classical parametric work, but new advances have more of non-parametric flavour. Intersting that all these new analysis lead to similar conclusion as from Fisher's LDA for the Iris data.
Overall, with computational power, the whole subject of Multivariate Analysis has changed its emphasis. Deriving sampling distributions as one of the topics which Fisher pioneered has now moved to simulation methods, for example, to obtain percentage points. Boot strapping is another innovation. Now the topic of High Dimensional Data is another growing area and Cluster Analysis has become the topic of unsupervised learning. Finally, we end with a historical note pointing out some correspondence between D'Arcy Thompson (pioneer of Shape Analysis) and R A Fisher where we could have seen Fisher’s insight into Shape Analysis but this collaborative work via a Research Student did not materialise.
Video recording of the 40th Fisher Memorial lecture.
|Fisher Memorial Trust Secretary Stephen Senn introducing Kanti Mardia
|A slide from the lecture
|Fisher Memorial Trust Chairman, Brian Charlesworth, presenting Kanti Mardia with the silver bowl commemorating his lecture
|Walter Bodmer thanking the Fisher lecturers
The 39th Fisher lecture was given on 12 July 2019 by Brian Cullis and Alison Smith at the 7th IBS Channel Network Conference , Rothamsted. The lecturers gave a spirited account of their increasing frustration at the lack of basic understanding of principles of experimental design and analysis shown by some life scientists and the practical steps they had taken through creation of their Design Tableau approach to providing a principled framework that life scientists could use to apply appropriate methods to design and analysis. This incorporated ideas that statisticians working at Rothamsted and elsewhere had developed, starting with Fisher himself. The 100th anniversary of Fisher's arrival at Rothamsted made a fitting occasion for a fine lecture that was appreciated by the audience in particular because the lecturers appropriately illustrated theoretical points with practical examples. Walter Bodmer, as Fisher Memorial Trust chairman, chaired the session and presented the speakers with the customary engraved sliver bowl afterwards. This was followed by lunch in honour of the lecturers, which lecturers, Trust members and various guests, including several Fisher family members, attended.
Fisher family members Jenny Tebboth, Rose Newsom and Sarah Posey
The title and abstract for the lecture are given below.
We were fortunate enough to have trained and worked as young biometricians when analysis of variance (ANOVA) techniques were the primary method of analysis for comparative experiments.
Our tool of trade was the GENSTAT package, so that the elegant notation of Wilkinson and Rogers and the framework of Block and Treatment structures became ingrained in our statistical thinking. So, despite the complexity of the LMM we now use, we appreciate the importance of maintaining these fundamental concepts, in particular the link between the analysis and the experimental design. We are concerned that this view is shared by only a few, as is evidenced by what we regard as a widespread mis-use of LMM for comparative experiments. This may either be due to an unintentional lapse in transitioning from ANOVA to LMM or a complete lack of exposure to traditional methods of analysis for comparative experiments.
Over recent years, we have made it a priority to fill in this gap for our young statistical colleagues at the University of Wollongong. In particular we have attempted to provide a link between ANOVA and LMM and to explain how to derive LMM that reflect the randomisation employed in the design of the experiment, no matter how complex. We found this to be a non-trivial task and tried numerous educational tools but without great success. A turning point was Brian's introduction of an Honours Statistics course on experimental design at the University of Wollongong. He based this course on Rosemary Bailey's book and found words of wisdom that have inspired us to develop an approach that we have termed Design Tableau (DT). The main aim of DT is to provide a simple, but general series of steps for specifying the LMM for a comparative experiment. It is founded on the seminal work of Sir Ronald Fisher, John Nelder, Rosemary Bailey and Robin Thompson. The motivation and concepts underlying Design Tableau will form the basis of our presentation. We will discuss the formal link between ANOVA and LMM, describe the steps that constitute DT, illustrate DT for simple cases in which the LMM may be used to re-produce an ANOVA and finally demonstrate how DT can be applied in a wide range of complex comparative experiments.
Fisher grandson Richard Newsom with Walter Bodmer
The 38th Fisher Memorial Lecture was given by by Professor Michael Goddard in Edinburgh on 9 October 2018 as part of a centenary meeting sponsored by the Fisher Memorial Trust, the Genetics Society, the Galton Institute, the London Mathematical Society and the Royal Statistical Society, to celebrate the publication of Fisher's famous paper: "The correlation between relatives on the supposition of Mendelian inheritance. Transactions of the Royal Society of Edinburgh. 1918;52:339-433.
Mike delivered a fine lecture as a fitting end to an excellent day in which Fisher's legacy and more modern developments were presented by a number of research veterans and some early career researchers. This was followed by a reception open to all attendees at which the Fisher Memorial Trust were delighted to have the presence of Fisher's grandson Richard Newsom. This was in turn followed by the usual Fisher Memorial Dinner for the lecturer and guests.
A simple but surprising result is that most quantitative genetic variation is due to very many polymorphisms, each with a tiny effect on the trait, and segregating in the population at moderate allele frequencies. For instance, there are approximately 10,000,000 sites in the genome where a mutation can affect a typical quantitative trait. Each generation, mutation generates new variation; some of the new mutant alleles have a large effect, but selection keeps them very rare. Most of the variation is caused by mutations of small effect that are almost neutral, and hence segregate at moderate allele frequencies. However, occasionally a mutation is favoured by selection and while it segregates generates a large variance. This is most common when the environment changes greatly so that the direction of selection on some mutations reverses. This new understanding explains many previously puzzling results, such as the linear response to artificial selection and the failure to find the genes causing variation in complex traits.
The ability to genotype thousands of SNPs at moderate cost has been utilised in methods of genomic selection or genomic prediction, which predict the genetic value of individuals for a trait based on SNP genotypes. This method is revolutionising animal and plant breeding and will be important in human medicine, for instance in personalised medicine.
Fisher pages last updated 10 February 2024
Page maintained by Stephen Senn email@example.com