spacer picture
spacer picture

Econometrics at the University of Illinois

spacer picture

Dictionary of Received Ideas of Statistics

This page is a fragment of a dictionary of received ideas of statistics. Most of the entries were composed one conference evening after too much wine. It is obviously modeled after Flaubert's famous compendium but inevitably is only the palest shadow of the work of Le Maître. As a restorative you can find a scanned version of Flaubert's Dictionary in English. For the original: en français.

Aristotle: Claimed you can fill space with regular tetrahedrons, but you can't. Don't mention horses teeth, or any Bacon.

Bayesianism: Like Marxism, better in theory than in practice.

Bayesians: Ask: How should one compute the discounted expected utility of one's afterlife? See Pascal's wager. With some trepidation, quote Maurice Kendall: "I have lamented that Bayesian statisticians do not stick closely enough to the pattern laid down by Bayes himself: if they would only do as he did and publish posthumously we should all be saved a lot of trouble." [On the Future of Statistics -- A Second Look, JRSS(A), (1968), 131, 182-204].

Casual Empiricism: cite Virgil, "ab uno disce omnes", from one example all is revealed.

Causal Inference: see Casual Empiricism.

Chance: Ask what does "a 40 percent chance of rain" mean in Iowa? Also mention the fabulous painting "Allegory of Fortune" by the 16th century Ferrarese master Dosso Dossi.

Coin-flipping: the real foundation of probability, viz. Rosencrantz and Guildenstern Are Dead. Consider also the grook:

A Psychological Tip

Piet Hein

Whenever you're called on to make up your mind,
and you're hampered by not having any,
the best way to solve the dilemma, you'll find,
is simply by spinning a penny.
No -- not so that chance shall decide the affair
while you're passively standing there moping;
but the moment the penny is up in the air,
you suddenly know what you're hoping.

Co-median: the median of a distribution, F, is usually defined, in accordance with cadlag conventions, as the infimum of the set of values, S = {x | F(x) >= 1/2}. This is the smallest value that minimizes E |X - mu|. When there are several minimizers of this expectation we may refer to the other (larger) ones as co-medians. [See Stigler (1977, Fractional Order Statistics), note added March 2013, I thought this was my joke, but it turns out I was just recycling Stigler.]

Confidence: What science wants, and the statistician lacks. Mention that EB Wilson (JASA, 1927) was the first to insist that random (confidence) intervals cover the true parameter with some probability, rather than that a random parameter lies in fixed interval with some probability. No less than Milton Friedman seemed confused about this point, see Neyman's (1938) Lectures and Conferences on Mathematical Statistics, p 145.

Education: Quote Deng Xiaopeng: "Execution is an indispensible means through which to educate the masses."

Econometrics: Mention Stefan Valavanis's (1959) textbook characterization: "Econometric theory is like an exquisitely balanced French recipe, spelling out precisely with how many turns to mix the sauce, how many carats of spice to add, and for how many milliseconds to bake the mixture at exactly 474 degrees of temperature. But when the statistical cook turns to raw materials, he finds that hearts of cactus fruit are unavailable, so he substitutes chunks of cantaloupe; where the recipe calls for vermicelli he uses shredded wheat; and he substitutes green garment dye for curry, ping-pong balls for turtle's eggs and, for Chalifougnac vintage 1883, a can of turpentine."

Endogoneity: Ask who put the dodgy in it.

Error: Cite Pareto: "Give me fruitful error anytime full of seeds, bursting with its own contradictions. You can keep your sterile truth for yourself."

ESP: Quote Mosteller "If there is ESP, that is exciting. However, thus far it does not look as if it will replace the telephone." (1991 Stat. Science)

Elicitation: see tuning parameter, mention Tom Sawyer's method of painting fences -- find someone else to do your dirty work.

Evidence: Quote the Venerable Sherlock Holmes: "I see no more than you but I have trained myself to notice what I see." Or Richard Pryor: "Who are you going to believe, me, or your lying eyes."

Finance: the alchemy of modern science.

Gambling: mention Aldous Huxley's comment, "...the relation between craps and Reality is a real one", letter of December 1969, cited in OED. Or quote Samuelson: "When I go to a casino, I go not alone for the dollar prizes but also for the pleasures of gaming -- for the soft lights and sweet music." (Econometrica, 1952, p 671).

Gauss: mention J. L. Bernstein's advice, "it is wise to degauss the heads prior to each recording session." Cited in the OED.

History: Statistics with small samples. Cite Mark Twain, "History doesn't often repeat itself, but it rhymes."

Iidly: as in "drawn iidly" --drawn as independent and identically distributed, that is somewhere between idly, and ideally, although we shouldn't discount the Freudian connotation "drawn by the Id." See independence.

Image Processing: Suggest the sphinx as an example of image denosing.

Inequality: The toolbox of mathematics is filled with them. Quote GH Hardy: "A science is said to be useful if its development tends to accentuate the existing inequalities in the distribution of wealth," (A Mathemetician's Apology, p 120), but note that he called this remark -- only an excusable rhetorical flourish of 1915.

Independence: The great superstition of probability theory. See Kac's Statistical Independence, and marvel at the apparent sophistication of the Haverford undergrads in the audience. See also iidly.

Kolmogorov: Thought that every single discovery should fit in a four-page Doklady note, since "the human brain is not capable of creating anything more complicated at one time." quoted in http://homepages.cwi.nl/~paulv/papers/kolmogorov.pdf

Literature: Statistics with falsified data.

Love: Brownian emotion.

Martingale: In France mention Rabelais' socks, citing the OED.

Median: is the message.

Moments: mention Tukey's "misty staircase," or say "quantiles are moments too, of a sort."

Normal: Mention Stigler's comment, "a rare one-word oxymoron". see also Gauss.

p-value: Mention the fictional treatise: Shame and the p-value. See evidence.

Pascal's Wager: quote Diderot, "Any Imam could just as well reason the same way."

Poker: quote Robert Pinsky quoting Walter Matthau (!) "Poker exemplifies the worst aspects of capitalism that have made our county so great."

Probability: Flip a coin for DeFinetti or Kolmogorov versions, see Chance. Quote Borel late in life: "Je vais pantoufler dans les probabilité", and if the verb isn't sufficiently clear quote Larousse: "Pour un fonctionnaire, en particulier pour un ancien élève des grandes écoles, quitter le service de l'État pour celui d'une entreprise privé."

R: what comes after Q. Quote Virginia Woolf (To the Lighthouse): "It was a splendid mind. For if thought is like the keyboard of a piano, divided into so many notes, or like the alphabet is ranged in twenty-six letters all in order, then his splendid mind had no sort of difficulty in running over those letters one by one, firmly and accurately, until it had reached the letter Q. He reached Q. Very few people in the whole of England reach the letter Q.... But after Q? What comes next?... Still, if he could reach R it would be something. Here at least was Q. He dug his heels in at Q. Q he was sure of. Q he could demonstrate. If Q then is Q--R--.... "Then R..." He braced himself. He clenched himself.... "...In that flash of darkness he heard people saying--he was a failure--that R was beyond him. He would never reach R. On to R, once more. R---.... "...He had not genius; he had no claim to that: but he had, or he might have had, the power to repeat every letter of the alphabet from A to Z accurately in order. Meanwhile, he stuck at Q. On then, on to R."

Regression: is demeaning.

Robustness: Burned with such intensity that, like Marxism-Leninism, only the ashes of the most pure remain.

Smoking: Say R.A. Fisher found it healthy and thereby became wealthy.

Statistics: The (futile) attempt to offer certainty about uncertainty. Cite Kafka, "I have reached the stage that I no longer desire certainty."

Structural estimation: see strictural estimation.

Strictural estimation: estimation subject to one or more strictures, cite OED -- " There is no strictural obstruction to the progress of the faeces." J.M. Duncan (1886).

Test: Quote "Joe" the CIA's chief "expert" on Iraq's aluminum tubes and their potential role for WMD, testifying in Congress on why he didn't consult Energy Department experts on the subject: "Because we funded it. It was our testing. We were trying to prove some things that we wanted to prove with the testing." (NYTimes, Oct 3, 2004). Mention that in sequential analysis this is called testing to a foregone conclusion.

Tukey: "never estimate intercepts, always estimate centercepts!" Quote von Neumann's comment: "There is this very bright graduate student, and the remarkable thing is that he does it all on milk." (Brillinger's (2002) obituary in the AMS Notices.) Tukey on Friedman: Milton was well acquainted with the statistical side and very sharp. Probably easier to cut yourself on him than anybody else in the Biometric Society or the Institute [of Advanced Study.]

Tuning parameter: What the Bayesian call the "degree of belief in the prior", say neither frequentists or Bayesians know how to select them.

Truth: see error. Ask: is it true that Somerset Maugham coined the the phrase "Nothing is too rum to be true"?

spacer picture

Econometrics
University of Illinois at Urbana-Champaign
Commerce West, Champaign IL 61820
contact Roger Koenker (roger@ysidro.econ.uiuc.edu)

spacer picture