November 11, 2013

Cautionary tales

Before the advent of cheap computing power, statistics was a rather dismal subject. David Lax scared me off from studying much of it by saying that 90% of statistics was done on sets of measure 0.

The following cautionary tale also dates to that era. Other light verse below. 

Hiawatha Designs an Experiment

Hiawatha, mighty hunter,
He could shoot ten arrows upward,
Shoot them with such strength and swiftness
That the last had left the bow-string
Ere the first to earth descended.

This was commonly regarded
As a feat of skill and cunning.
Several sarcastic spirits
Pointed out to him, however,
That it might be much more useful
If he sometimes hit the target.
“Why not shoot a little straighter
And employ a smaller sample?”

Hiawatha, who at college
Majored in applied statistics,
Consequently felt entitled
To instruct his fellow man in
Any subject whatsoever,
Waxed exceedingly indignant,

Talked about the law of errors,
Talked about truncated normals,
Talked of loss of information,
Talked about his lack of bias,
Pointed out that (in the long run)
Independent observations,
Even though they missed the target,
Had an average point of impact
Very near the spot he aimed at,
With the possible exception
of a set of measure zero.

“This,” they said, “was rather doubtful;
Anyway it didn’t matter.
What resulted in the long run:
Either he must hit the target
Much more often than at present,
Or himself would have to pay for
All the arrows he had wasted.”

Hiawatha, in a temper,
Quoted parts of R. A. Fisher,
Quoted Yates and quoted Finney,
Quoted reams of Oscar Kempthorne,
Quoted Anderson and Bancroft
(practically in extenso)
Trying to impress upon them
That what actually mattered
Was to estimate the error.
Several of them admitted:
“Such a thing might have its uses;
Still,” they said, “he would do better
If he shot a little straighter.”

Hiawatha, to convince them,
Organized a shooting contest.
Laid out in the proper manner
Of designs experimental
Recommended in the textbooks,
Mainly used for tasting tea
(but sometimes used in other cases)
Used factorial arrangements
And the theory of Galois,
Got a nicely balanced layout
And successfully confounded
Second order interactions.

All the other tribal marksmen,
Ignorant benighted creatures
Of experimental setups,
Used their time of preparation
Putting in a lot of practice
Merely shooting at the target.

Thus it happened in the contest
That their scores were most impressive
With one solitary exception.
This, I hate to have to say it,
Was the score of Hiawatha,
Who as usual shot his arrows,
Shot them with great strength and swiftness,
Managing to be unbiased,
Not however with a salvo
Managing to hit the target.

“There!” they said to Hiawatha,
“That is what we all expected.”
Hiawatha, nothing daunted,
Called for pen and called for paper.
But analysis of variance
Finally produced the figures
Showing beyond all peradventure,
Everybody else was biased.
And the variance components
Did not differ from each other’s,
Or from Hiawatha’s.

(This last point, it might be mentioned,
Would have been much more convincing
If he hadn’t been compelled to
Estimate his own components
From experimental plots on
Which the values all were missing.)

Still they couldn’t understand it,
So they couldn’t raise objections.
(Which is what so often happens
with analysis of variance.)
All the same his fellow tribesmen,
Ignorant benighted heathens,
Took away his bow and arrows,
Said that though my Hiawatha
Was a brilliant statistician,
He was useless as a bowman.
As for variance components
Several of the more outspoken
Make primeval observations
Hurtful of the finer feelings
Even of the statistician.

In a corner of the forest
Sits alone my Hiawatha
Permanently cogitating
On the normal law of errors.
Wondering in idle moments
If perhaps increased precision
Might perhaps be sometimes better
Even at the cost of bias,
If one could thereby now and then
Register upon a target.

— W. E. Mientka, “Professor Leo Moser – Reflections of a Visit” American Mathematical Monthly, Vol. 79, Number 6 (June-July, 1972)

That link also has a number of limericks, not all original. I particularly like the egregious set theoretic pun, on the occasion of a Bertrand Russell guest lecture, in:

Condemned for defending the masses
Scourged for defaming the lasses
Not moved by disgrace
He has come to this place
To teach the class of all classes.

And I’ve always loved:

Nature and nature’s law lay hid by night
Then God said “Let Newton be,” and all was light.

— Alexander Pope

It did not last; the Devil shouting “Ho!
Let Einstein be!” restored the status quo.

— John Collings Squire

Comments

One Response to “Cautionary tales”

  1. Keshav Murthy on November 12th, 2013 1:17 pm

    Good Change of pace from all the prose. I happen to like the Pope & Squire poems.

    When we released Informix Warehouse Accelerator based on the BLINK research, I used it as a template to write this…http://ibm.co/18oW8dH

Leave a Reply




Feed: DBMS (database management system), DW (data warehousing), BI (business intelligence), and analytics technology Subscribe to the Monash Research feed via RSS or email:

Login

Search our blogs and white papers

Monash Research blogs

User consulting

Building a short list? Refining your strategic plan? We can help.

Vendor advisory

We tell vendors what's happening -- and, more important, what they should do about it.

Monash Research highlights

Learn about white papers, webcasts, and blog highlights, by RSS or email.