Richard D. Gill's home page
Mathematical Statistics
(Emeritus professor)
You can read my "farewell lecture" (September, 2017) at
my blog
"
To contact me quickly, try email
(surname at math dot leidenuniv dot nl), or mobile phone.
Click here for my postal and visiting addresses,
mobile phone number, various other email addresses and other contact possibilities.
In particular, I have started a one man statistical consultation business
Combray Causality Consulting.
Every statistician in the world is currently working on the Corona crisis.
I have been looking at the results of the Marseilles trial of
hydroxychloroquine and azithromycin in very early Covid-19 infection. I've also
been studying the data of Dutch GP Rob Elens. A conference talk for the
Bernoulli-IMS One World Symposium (August 2020) is in preparation.
Here is a preview of the slides of my talk Presentation.pdf
(recently updated) and here is the
video presentation (youtube video).
Here is a short version of the slides
(for a three min "elevator pitch"). Here is a
short intro to who I am.
Here are some preliminary analyses with R and with JASP:
https://rpubs.com/gill1109/raoult,
raoultJASP.html. More to come, soon.
And a recent talk "Corona statistics, Corona uncertainty"
slides (pdf),
video of part 2; open discussion.
Part 1 of the video to be added, soon.
See also Rob Elens data analysed
and open letter to NEJM.
A Zoom commemoration of W.R. van Zwet (207 MB mp4;
duration 35 minutes; 5 minutes silence between ca. 25 - 30 min. when Zoom briefly fails us).
Photographs collected by many friends of Willem:
Blog posting, "RIP Bill van Zwet"".
I have revised and extended my farewell lecture
"From killer nurses to quantum entanglement",
and am slowly converting it into a new paper
"Statistical issues in Serial Killer Nurse cases"
with amazing new statistical graphics on the (UK) Ben Geen case.
Latest version: 14 February 2020. Just some appendices still to be written.
See also my recent Ben Geen talk. Here are links to
the expert opinion of
Jane Hutton and the
expert opinion of
Professor David Denison
(who sadly died in 2014)
at Ben's appeal in 2009 concerning the statistical, or rather scientific issues,
around "unexpected clusters of medical cases" and their association with
particular health care workers.
The judge did not allow Jane to summarise her "Opinion" (technical term for an
expert's report) in court in person because it was "merely common sense",
as was also Denison's Opinion. At the judge's summing up, the jurors were
instructed to ignore that evidence. No common sense in my court, please! (shades
of Sally Clark: it was "not rocket science"). This obviously unfair legal ruling
provided Ben's legal team with legal grounds for approaching the CCRC
and hiring me: I was to perform the epidemiological
research which Jane had argued needed to be performed.
I'm a member of the North Sea Group,
a circle of scholars and scientists mainly from countries around the North Sea,
interested in legal evidence and legal proof.
We have a webpage with blog and
forum.
I'm webmaster.
The group has been meeting at annual workshops already for several years.
A speciality is to study specific cases in depth using
different methodologies from computer science and artificial intelligence,
from psychology, philosophy and law, from statistics and probability.
"
Some thoughts on Bell’s theorem & on Bell-denialism,
Zoom.us talk at QM foundations – nature of time seminar, J U Krakow, 5:00 p.m. Warsaw time, April 6, 2020.
Krakow talk, continued,
Zoom.us talk at QM foundations – nature of time seminar, J U Krakow, 5:00 p.m. Warsaw time, April 13, 2020.
Blog instalment "Warsaw"
Yet Another Statistical Analysis of the data of the (2015) "Loophole-Free" Bell-CHSH Experiments,
Vaxjo "QIRIF" conference, June 2019.
Lecture course "The Role of Statistical Science
in Quantum Information", given at Fudan University, Shanghai,
and Peking University, Beijing, September/October 2019.
Illustrations: the satellite Micius was used in Jianwei Pan's
Earth (Shanghai) - Space (Micius) - Earth (Beijing) quantum key-distribution
experiment in 2017. Here are a links to directories containing materials for the courses at
Fudan University and at Peking University.
I will continue to organise this material and perhaps write a small monograph in
"lecture notes" style.
Foraging for wild mushrooms.
The intersection axiom of conditional independence.
Dutch New Herring, correlation vs causation, good scientific research practices.
It is getting hard to find the two Tilburg University press releases
and economist Ben Vollaard's two "AD Haringtest" papers on internet, so here are links:
AD Haringtest paper July 2017,
AD Haringtest update November 2017,
Press Release July 2017,
Press Release November 2017.
I will put the links into the slides, as well as update with my latest research findings
on time-dependence through self-selection of participants and panel: the "samples" are
not random samples.
I had a conflict of interest - I was asked in 2018 for my scientific opinion on the
original papers by one of the parties in the ensuing legal (scientific integrity) case.
The original documents are all in Dutch. For three years, Dr. Vollaard declined to
share the data-sets which he analysed. The Dutch national institution for dealing with
alleged failures of scientific integrity (LOWI) has in October 2019 (advice no. 23)
concluded its work on the case. The conclusion can be found (anonymised) on their website.
LOWI writes "The LOWI would applaud it if a scientific debate could take place
between all parties involved, and experts". They also write "It is clear that mistakes
were made in the press release which presented the results of this research
to the public". The data-sets can at last be downloaded from Dr. Vollaard's personal
webpage. Unfortunately they are in the proprietary format of the commercial "Stata"
package, which I do not wish to pay for. Fortunately, Stata data files can now be fairly
easily read into R and converted to non-proprietary formats.
I can email a csv file to anyone interested.
26 August, 2015. They did it at last! A first reaction.
What did they do, exactly, and what was my (modest) scientific
contribution.
Illustrations: Figure 7 from "Bertlmann's socks and the nature of
reality", Bell (1981); and the set-up in Delft, 2015.
See "Bell games" for a comparison of the standard Bell
experiment and the Delft "event-ready detectors" version built on top of
entanglement swapping.
Interview with John Bell (OMNI, 1988)
Quantum Love (Delft YouTube video
I must admit to getting a lot of fun, and scientific stimulation, from
interacting with Bell-deniers. I apologise for the names I use. It can be argued that
such derogatory terms do not belong in civilised public scientific conversation,
but unfortunately they are common in private conversation,
which of course is also part of science. I feel that those who suspect that such labels
are being applied to them should proudly wear them
as badges of honour. As Gandhi said: first they ignore you, then they fight you,
then you win. When they call you names, it means you are somehow a threat to them.
It might mean that they are losing the argument.
In fact, I consider this an important part of science-outreach:
how can we explain Bell's theorem to the general public?
Well, a good start is to try to understand the mentality of very smart and
well-educated people who believe that Bell made some fundamental but simple
mistake, that they have exposed that mistake, and that there is an establishment
conspiracy to suppress their findings. I believe that Bell's argument is very simple
and absolutely correct, but apparently it is hard to get across.
Part of the problem is, I believe, the fact that it is impossible to
"understand" quantum mechanics, in the sense that we usually use the word
"understand". In fact, that is exactly the content of Bell's theorem. A "local realistic"
interpretation of quantum mechanics is precisely:
an interpretation that we can understand. We can get used to the mathematics, we can
even gain intuition about the the *mathematical world* which the conventional rules of
quantum mechanics describes. But those rules will never make sense as a picture of how
the world actually operates.
The images below come from a new version of a paper I wrote a few years ago entitled
"The triangle wave versus the cosine".
The paper studies the "spinning coloured disk model" of the so-called EPR-B correlations.
It is about the class of all correlation functions
which classical physics can generate, in a situation when we expect certain
symmetries and "certainty relations". The paper describes many open problems in classical
probability theory, some inspired by looking at computer simulations.
Here's my latest work in this field - a response to a
recent paper by Joy Christian
published in the journal Royal Society Open Science. I have submitted my comment
to the same journal. And here is a draft of a new paper
on Steve Gull's sketch of a proof
of Bell's theorem using Fourier analysis. Here are
slides of a talk on same subject. Here's
something on Joy Christian's recent work in IEEE Access.
I was asked by defense lawyers
working on behalf of Ben Geen, to analyse statistical data of
occurrence of respiratory arrest in accident and emergency at numerous
UK hospitals
(draft R html notebook,
draft report).
After I had done my statistical work, I dug more deeply into
other aspects of the case. The more I discovered, the more I was shocked that
this case was a carbon copy of our own (Netherlands) case of Lucia de Berk.
However there are three important differences. (1) The UK media did an even more
perfect hatchet job on him than the Dutch media did on Lucia. (2) The Criminal Cases
Review Commission is understaffed and underfunded, and takes a formalistic (legalist)
view: give us a "new legal fact" or we will do nothing. (3) There is no Metta de Noo:
no medical whistleblower. In the Lucia case, Metta de Noo fought for 7 years to get
Lucia the fair (re-) trial which Lucia deserved. Metta is a senior medical expert,
well connected in society, and she had inside information about the case.
In the Lucia case, what was needed (and finally happened at the re-trial in Arnhem,
2009--2010) was an independent and thorough re-appraisal of
existing (medical) evidence.
Ben has by now sat out about 12 years of his 30 year sentence.
Because he claims to be innocent
he is denied any "good prisoner" benefits and
will also never get parole or early release.
A small supporters' group set up a website
Justice for Ben Geen.
Some UK journalists, at "The Times" no less,
briefly showed some interest:
Nurse 'was victim of Shipman hysteria'
For a long time nothing has happened but right now, things are moving again.
YouTube video. Slides of my talk
Murder by Numbers.
The naked truth about the case of Lucia de B.
A historical document: statement by Haga-Hospital, 2010,
regarding the acquittal of Lucia de Berk:
English translation,
original.
Four days after the TEDx event, I saw the movie
Lucia de B.
on its premiere night in Amsterdam. Here is my personal
film review: Splendid acting, very moving, beautifully told human story,
centering around Lucia herself. Despite compression of the story line and focus
on Lucia's personal experiences, it still contained such key features as:
the
close personal links
between key people from hospital, justice and experts (image right);
the mental illness and mental breakdown of the chief-paediatrician at JKZ ...
There was a vain and ambitious hospital director. A bad statistician.
Real life heroine Metta de Noo and hero Ton Derksen
were concentrated in the film into the imaginary person
of one imaginary whistleblower at the last place you might expect
to find them: in the Public Prosecution service. But on the other hand:
it wasn't black and white. There were good medics and bad medics,
good nurses and bad nurses, good cops and bad cops ... Apparently,
even some people in the Public Prosecution service found the witch hunt
deeply disturbing.
For more (much, much more) on the Lucia case,
see several more items below, as well as yet more
items from past versions of my home-page:
writings on Lucia.
R-fun
Some years ago
I offered a prize for the person who remasters the logo of the
VVS: the Dutch statistical society
(top image) in the most beautiful postscript. An exercise in
curve fitting with splines, perhaps?
Better still would be a mathematical/statistical story of the curves
themselves, providing an elegant parametric family which reproduces the
whole logo. Finally I decided to do it myself, and I think I am getting
close with this perspective image of some very simple 3-dimensional
curves, with indeed a statistical story behind them
(bottom).
The R script which draws this logo can be found here.
It should generate a rotatable 3-d image...
See the slides of my
Amst-R-dam R users group meetup 2011
(updated 2012) talk
R-fun: part 1, the VVS logo in R; part 2, R on an iDevice.
For some old news on "R on an iDevice" see the 2014 talk
R on an iDevice,
given at a Data Science NL meetup.
For more R fun: I am nowadays an enthusiastic user of RStudio and RPubs.
You can find all kinds of
R work by me at my RPubs site.
VVS stands for "Vereniging voor Statistiek". SMS stands for
"Section Mathematical Statistics".
The VVS also has an OR section, hence the common alternative name VVS-OR. And nowadays
the society has a nice new logo, though it doesn't look to me so much
like anything to do with statistics
Teaching (already seems like long ago!)
Spring 2016
FSG: Tuesdays 13:45--15:30 room 405
Master's level (or advanced Bachelor's)
The master specialization Statistical
Science for the Life and Behavioural Sciences is a collaboration of our group with
others in biomedical statistics, biostatistics, and psychometrics.
Here you can find links to various courses I have given in the past,
in particular quantum statistics,
statistics for astronomers,
HOVO courses (adult education courses, in Dutch) on
use and abuse of statistics, forensic science
(Hovo-criminalistiek-statistiek-1,
Hovo-criminalistiek-statistiek-2,
Hovo-criminalistiek-statistiek-3).
Research
For some outdated impresions of my research interests,
take a look at various talks on
Slideshare.
My last PhD student was
Giulia Cereda,
working on forensic statistics,
in particular the rare haplotype problem (aka the "fundamental problem
of forensic mathematics"). Giulia was in a "cotutelle" arrangement:
this was a joint project between Lausanne (Franco Taroni) and Leiden.
Interests (somewhat out of date), most active marked *
Not so recent talks and papers
Publications