Where there is no uncertainty , there cannot be truth — Richard Feynman


    • Phone: 416-736-2100 ext. 77164
      Fax: 416-736-5757
    • Office: North 626 Ross Building
    • Office Hours: Fridays: 8 am to 10 am (January to early April 2017)
    • e-mail:
      Georges dot Monette PLUS hp AT mathstat.yorku.ca

What plot do you (almost) see twice in each rotation?

Added variable plot

See the bottom of the page for an answer.


... in all the sciences, we’ve got to make sure that we are supporting the idea that they’re not subject to politics, that they’re not skewed by an agenda, that, as I said before, we make sure that we go where the evidence leads us.” – Barack Obama

Often decisions about interventions must be made, even if based on limited empirical evidence, and we should help decision-makers make sensible decisions under clearly stated assumptions so that "consumers" of the conclusion about the effects of some intervention can honestly weigh the support for that conclusion. -- Donald Rubin

Human History becomes more and more a race between education and catastrophe. – H. G. Wells

The rubber hits the road where the data hits the code. – Janet McDougall

The best thing about being a statistician is that you get to play in everyone's backyard. – John W. Tukey,who, incidentally, coined the terms 'software' and 'bit'. 

Once you know hierarchies exist, you see them everywhere – Ita Kreft and Jan de Leeuw (1998) “Introducing Multilevel Modeling”

Once you tune into ellipses, you will begin to see them everywhere ... – James McMullan

The best thing about being a statistician is that you get a license to poke your nose into everyone else's business. – (??)

The best thing about universities is individual freedom, the worst thing is collective irresponsibility. – David Northrup

Humanists believe that the world has a fixed number of mysteries, so that when one is solved, our sense of wonder is diminished. Scientists believe that the world has endless mysteries, so that when one is solved, there are always new ones to ponder. – D. O. Hebb quoted by Steven Pinker

Far better an approximate answer to the right question, which is often vague, than an exact answer to the wrong question, which can always be made precise. – John W. Tukey, (1962), “The future of data analysis.” Annals of Mathematical Statistics 33, 1-67.

A bad answer to a good question may be far better than a good answer to a bad question. – a graduate class extrapolating from Tukey's dictum.

It is better to know some of the questions than all of the answers. – James Thurber

I have a lot of questions . . . but I don't know what they are. – overheard at the end of what must have been an inspiring lecture

The worst, i.e., most dangerous, feature of  'accepting the null hypothesis' is the giving up of explicit uncertainty . . . Mathematics can sometimes be put in such black-and-white terms, but our knowledge or belief about the external world never can. – John W. Tukey. (1991). “The Philosophy of Multiple Comparisons.” Statistical Science 6, 100--116.

Where there is no uncertainty there cannot be truth. – Richard Feynman (confirmed by Bill Langford)

So far as the theories of mathematics are about reality, they are not certain; so far as they are certain, they are not about reality. – Albert Einstein

Doubt is not a pleasant condition, but certainty is absurd.

Art is a lie that enables us to realize the truth. – Pablo Picasso

At their best, graphics are instruments for reasoning. – Edward Tufte,  www.edwardtufte.com

An elementary demonstration is one that requires no knowledge – just an infinite amount of intelligence. -- Richard Feynman.

Box and Cox transformedAll models are wrong but some are useful. – George E. P. Box

All models are wrong but, we hope, not as wrong as the ones we used earlier. –paraphrased from Isaac Asimov

'All models are wrong' is a model, thus must be wrong. Perhaps it's useful - paraphrased from a comment on a blog.

The business of the statistician is to catalyze the scientific learning process. – George E. P. Box

There are no routine statistical question; only questionable statistical routines. – D.R. Cox

We at York must give special emphasis to the humanizing of man, freeing him from those pressures which mechanize the mind, which make for routine thinking, which divorce thinking and feeling, which permit custom to dominate intelligence, which freeze awareness of the human spirit and its possibilities... – Murray G. Ross

It is better to be wrong than to be vague. – Freeman Dyson

It is much more important to be clear than to be correct. – Blair Wheaton

Science may be described as the art of systematic over-simplification. – Karl Popper

There are three kinds of lies: lies, damned lies, and statistics. – Mark Twain with attribution to Benjamin Disraeli

Lies--damned lies--and statistics – Leonard Henry Courtney with attribution to a “Wise Statesman,” possibly Disraeli [see http://www.york.ac.uk/depts/maths/histstat/lies.htm]

Statistical thinking will one day be as necessary for efficient citizenship as the ability to read and write. – S. Wilks attributing to H. G. Wells

A certain elementary training in statistical method is becoming as necessary for anyone living in this world of today as reading and writing. –  H. G. Wells in "The Informative Content of Education," The Presidential Address to the Educational Science Section of the British Association for the Advancement of Science, given on September 2nd, 1937.

Statistical literacy is a necessary precondition for an
educated citizenship in a technological democracy – Gerd Gigenrenzer et al.

It is easy to lie with statistics. It is hard to tell the truth without it. – Andrejs Dunkels

Data analysis is an aid to thinking and not a replacement for. – Richard Shillington

Methodology should never be allowed to displace intelligence. – [paraphrased from Lee Wilkinson, I think]

Another thing about fit indices is that they are used all too often as substitutes for thinking. In most cases, statistical analysis should be not about determining the ”best fitting” model according to a single numerical criterion. In any given research there hopefully are underlying substantive theory and knowledge, the research hopefully is guided by research questions and knowledge about control variables, there is a distinction between primary and secondary interest, a single research often has elements of hypothesis testing as well as exploration, etc. etc. Fit indices in the ?IC family are useful only as a secondary type of summary information, but research questions and existing  knowledge are more important.  – Tom Snijders

If you try to estimate everything, you will end up estimating nothing. – [I forget who said this but I'd like to know!]

Fishing for hypotheses is like throwing a dart at a wall and then drawing a target around it. –  Andrée Monette

When statistics are not based on strictly accurate calculations, they mislead instead of guide. The mind easily lets itself be taken in by the false appearance of exactitude which statistics retain in their mistakes, and confidently adopts errors clothed in the form of mathematical truth. – Alexis de Tocqueville [With the benefit of a few centuries to reflect on this, we appreciate that the accuracy of the calculations is only one of many requirements to ensure that statistics guide and not mislead]

Causal interpretation of the results of regression analysis of observational data is a risky business.  The responsibility rests entirely on the shoulders of the researcher, because the shoulders of the statistical technique cannot carry such strong inferences. – Jan de Leeuw.

Correlation is not causation - but it sure helps– Edward Tufte

Correlation does not imply causation but it does waggle its eyebrows suggestively and gesture furtively while mouthing 'look over there. – Randall Munroe, xkcd.com.

OK! Correlation does not imply causation yada yada. – Paul Krugman

Do I love you because you’re beautiful, or are you beautiful because I love you?  – Prince Topher to Cinderella

... a primary objective in the design and analysis of observational studies is to control, through sampling and statistical adjustment, the possible biasing effects of those confounding variables that can be measured: a primary objective of in the evaluation of observational studies is to speculate about the remaining biasing effects of those confounding variables that cannot be [or: were not] measured. – Donald B Rubin (Matched Sampling for Causal Effects, 2006)

Not everything that can be counted counts, and not everything that counts can be counted. – Albert Einstein

If a man is offered a fact which goes against his instincts, he will scrutinize it closely, and unless the evidence is overwhelming, he will refuse to believe it. If, on the other hand, he is offered something which affords a reason for acting in accordance to his instincts, he will accept it even on the slightest evidence.  –Bertrand Russell

There's no crime in being ignorant. Problems arise when people who don't know they're ignorant rise to power. -- Neil deGrasse Tyson (with thanks to Jen Agg)

It ain't what they don't know that's the problem – it's what they know that ain't so -- said of members of the U.S. Congress by ?? (communicated by David Brillinger).  Variants aimed at different groups are attributed to sources ranging from Will Rogers to Ronald Reagan. 

Moral indignation is jealousy wearing a halo. – H. G. Wells  [what does this have to do with statistics?]

If you can't be a good example, be a horrible warning. -- scrawled in a country bathroom.

From the Globe & Mail, Social Studies column by Michael Kesterton, September 9, 2003:

Random: Washington-area teenagers have been overheard saying such things as: ”Did you see that outfit she was wearing? That was so random!” “Who invited those random kids to this party?” “I never watch the news on TV. It's too, like, random.” The adjective seems to mean “serendipitous,” but is more value-neutral. “It's actually rather specific the way students use it,” English teacher Patrick Welsh tells The Washington Post, adding “the brightest of the bright kids are the ones who tend to use it.”

I have a soft spot for secret passageways, bookshelves that open into silence, staircases that go down into a void, and hidden safes. I even have one myself, but I won't tell you where. At the other end of the spectrum are statistics which I hate with all my heart. – Luis Buñuel

I believe that the artist doesn't know what he does. I attach even more importance to the spectator than to the artist. – Marcel Duchamp

Ellipse and conjugate axes generator

-- Marcel Duchamp

What difference does it make to the dead, the orphans and the homeless, whether the mad destruction is wrought under the name of totalitarianism or the holy name of liberty or democracy? – Mahatma Gandhi (1869 - 1948), Non-Violence in Peace and War

No problem is so big or so complicated that it can't be run away from. – Linus van Pelt ( Peanuts )

Natural Selection is a mechanism for generating an exceedingly high degree of improbability. – Sir Ronald Fisher

In times of change learners inherit the earth, while the learned find themselves beautifully equipped to deal with a world that no longer exists. – Eric Hoffer

Being a statistician means never having to say you're certain – ??

There'll be a time when not having strong opinions about anything will be seen not as a character flaw, but as a virtue. – Alberto Cairo

It has often and confidently been asserted, that man's origin can never be known: but ignorance more frequently begets confidence than does knowledge: it is those who know little, and not those who know much, who so positively assert that this or that problem will never be solved by science. – Charles Darwin

One of the painful things about our time is that those who feel certainty are stupid, and those with any imagination and understanding are filled with doubt and indecision. – Bertrand Russell

The methods of statistics turn art into science – paraphrased from Arnold Zellner

Statistics is an art struggling to be a science. –  Heather Krause 

A data scientist is a statistician who lives in San Francisco – ?

A data scientist is a statistician who is useful – Hadley Wickham

The absence of evidence is not evidence of absence – Carl Sagan and many many others cautioning against concluding that the null hypothesis is correct when you merely fail to reject it.

The absence of evidence is not evidence of absence – Donald Rumsfeld to George W. Bush justifying the attack on Iraq! -- and using it with the opposite intent to Sagan's!

A difference of evidence is not necessarily evidence of a difference – ??

If it were a fact, it wouldn't be called intelligence — Donald Rumsfeld
Wow! — Stephen Colbert

Changing your mind is the only sure proof you can offer that you've got one —  Richard P. Feynman quoting ??

Ignorance more frequently begets confidence than does knowledge. — Charles Darwin, The Descent of Man

Statisticians learn not to be surprised by the improbable – which is usually probable – only by the improbably improbable – ??

One might perchance say this was probable that things improbable oft will hap to men — Aristotle quoting Agathon

Railing against collinearity is like railing against gravity – anonymous referee commenting on an article on collinearity and variance inflation.

There once was a student of yore
Whose inference truly was poor.
From a sample of one,
His mean was .1,
And the variance he found was .4.
– G. Eric Moorhouse

If you think you understand X that's a sure sign that you don't understand X  – a metaquote.

Believe those who are seeking the truth. Doubt those who find it. – André Gide

If you amplify everything, you hear nothing. – Jon Stewart

Seek the company of those who seek the truth, and run away from those who have found it. – Vaclav Havel

The scientist is not a person who gives the right answers, he is one who asks the right questions. – Claude Lévi-Strauss (Le Cru et le Cuit, 1964) (*)

We are inclined to believe those whom we do not know because they have never deceived us. – Samuel Johnson

“Faith” is a fine invention
When Gentlemen can see —
But Microscopes are prudent
In an Emergency
– Emily Dickinson

Let us change our traditional attitude to the construction of programs: Instead of imagining that our main task is to instruct a computer what to do, let us concentrate rather on explaining to humans what we want the computer to do.
Donald E. Knuth, 1984(*)

...academic administrative positions must be about both leadership and management because one without the other leads to no results or to trivial results. – Sheila Embleton. For a few reflections on academic governance that are both deep and practical see a Y File article.

It's foie. You've got to get it right  – the “King of Ginger” rejecting a dish performing quality control at The Black Hoof

There's no task so impressive that it can't be ruined by a rubric – Hans Krause

It's not the data alone, but analytics – and people trained to use them – that generate real value from big data. – Suzanne Gordon, CIO, SAS

After Eisenhower, you couldn't win an election without radio. After JFK, you couldn't win an election without television. After Obama, you couldn't win an election without social networking. I predict that in 2012, you won't be able to win an election without big data. – Alistair Croll

RNum2Factor– R

The real battlefront is not between Islam and West. The real battlefront is between all the faith traditions . . . atheists among that . . . all the moderates against extremists. – Imam Feisal Abdul Rauf

This is not a scientific survey. It's a random survey.
– Representative Daniel Webster voting in May 2012 for the abolition of the American Community Survey, the U.S. analogue of the Canadian “long form."

There was a young man of Lyon
Who normally fished on the Rhône.
One hour he caught seven.
Next five. Then eleven.
Not normal!  Those fish were Poisson.
– G. Eric Moorhouse

90% of the world’s data was generated in the last two years and 80% of that data is unstructured ... – Geeknet, Inc. 2012

Statistical rituals largely eliminate statistical thinking in the social sciences. Rituals are indispensable for identification with social groups, but they should be the subject rather than the procedure of science. – Gerd Gigerenzer

... statistics is fraught with contextual issues, which is the nature of the discipline, whereas often mathematics strips off the context in order to abstract and generalize. – J. Michael Shaughnessy

I've worked in so many areas—I'm sort of a dilettante. Basically, I'm not interested in doing research and I never have been. I'm interested in understanding, which is quite a different thing. And often to understand something you have to work it out yourself because no one else has done it. – David Blackwell

A popular joke is that  "a data scientist is a statistician who lives in San Francisco," but Hadley Wickham, a Ph. D. statistician, floated a more cynical take on Twitter: "a data scientist is a statistician who is useful." Statisticians are the guardians of statistical inference, and it is our responsibility to educate practitioners about using models appropriately, and the hazards of ignoring model assumptions when making inferences. But many model assumptions are only truly met under idealized conditions, and thus, as Box eloquently argued, one must think carefully about when statistical inferences are valid. When they are not, statisticians are caught in the awkward position, as Wickham suggests, of always saying "no". This position can be dissatisfying.  – Ben Baumer in "A Data Science Course for Undergraduates: Thinking with Data", March, 2015

What plot do you (almost) see twice in each rotation?

The blue plane, which is kept almost on edge as the points rotate, is the graph of the simple regression of Health on Weight. The ellipse in the horizontal plane is the data ellipse for the two predictors, Weight and Height. The blue line in the horizontal plane is the regression line predicting Height from Weight.  The yellow plane is the graph of the multiple regression of Health on both Height and Weight.Twice in every rotation, we look down the intersection of the blue plane and yellow plane. At that same moment, the blue line in the horizontal plane points directly at us.  A linear shear transformation of the blue plane can make it horizontal.  Then the  horizontal displacement of points, to the right or the left of the vertical plane that goes through the blue line and the intersection of the blue plane and the yellow plane, is proportional to the residual in the regression of Height on Weight.  The vertical displacement of points above and below the blue plane is the residual of Health on Weight. Thus, we are looking at the Added Variable Plot (or the Partial Residual Leverage Plot) for the addition of Height to Weight in predicting Health.
So the short answer is: an Added Variable Plot.