Why model?

The first question that arises frequently--sometimes innocently and sometimes not--is simply, "Why model?" Imagining a rhetorical (non-innocent) inquisitor, my favorite retort is, "You are a modeler." Anyone who ventures a projection, or imagines how a social dynamic--an epidemic, war, or migration--would unfold is running some model.
But typically, it is an implicit model in which the assumptions are hidden, their internal consistency is untested, their logical consequences are unknown, and their relation to data is unknown. But, when you close your eyes and imagine an epidemic spreading, or any other social dynamic, you are running some model or other. It is just an implicit model that you haven't written down.

This being the case, I am always amused when these same people challenge me with the question, "Can you validate your model?" The appropriate retort, of course, is, "Can you validate yours?"  
The choice is not whether to build models; it's whether to build explicit ones. 

Here's more Joshua Epstein on modelling.

by datacharmer | Saturday, July 21, 2012
| 0 comments | | Why model? @bluematterblogtwitter

Lady tasting tea - The adventures of R.A. Fisher in statistical science

Already, quite soon after he had come to Rothamstead, his presence had transformed one commonplace tea time to an historic event. It happened one afternoon when he drew a cup of tea from the urn and offered it to the lady beside him, Dr. B. Muriel Bristol, an algologist. She declined it, stating that she preferred a cup into which the milk had been poured first. “Nonsense,” returned Fisher, smiling, “Surely it makes no difference.” But she maintained, with emphasis, that of course it did. From just behind, a voice suggested, “Let’s test her.”

Fascinating anecdote from the life of R.A. Fisher. The link has more on Fisher's exact test and on whether the lady was indeed able to tell the difference between cups.

Shopping in North Korea

by datacharmer | Saturday, March 31, 2012
  | 1 comments | | Shopping in North Korea @bluematterblogtwitter

The politics of hate, Greece edition

The moral condemnation directed towards the Greeks is the saddest aspect of this financial crisis, and it says a lot more about the givers than the receivers. Just like the smell of burning books, the demonisation of entire populations and a 'we good, they bad' mentality is an unmistakeable manifestation of pathology. Lacking the intelligence or the interest to understand what the hell is going on, the morally outraged resort to silly stories and juvenile good/evil dichotomies to make sense of the world.

Richard Parker, the Harvard economist, has an interesting article in the FT:

Greece desperately needs reforms. Mr Papandreou knew this well before the crisis, as did most Greeks. But the larger problem was the panic that swept over Europe and the easy moralising that financial crises evoke. Greeks were cast as tax-evaders, lazy and anti-business, their government as over-indebted, bloated and corrupt – a situation that required castor oil and humiliation.

But almost none of the moralising clich├ęs were true. Greek taxes were more than a third of gross domestic product, near the European average. And if Greeks were anti-business, why then were there more small entrepreneurs per capita than anywhere else in Europe? Government was not bloated in terms of employees – at a fifth of the labour force, it was about the European average. Corruption was clearly a problem, but our data showed it was concentrated – incomprehensibly to non-Greeks – in the health sector, where minor “gifts” to doctors secured early scheduling of surgeries.

Two more facts to add here. One, Greeks work amongst the highest number of hours of all OECD countries. Two, while there is a lot of tax evasion in Greece (together with high tax rates so that overall tax take is comparable to elsewhere), this has to do with the fact Greece has a relatively large number of small, family businesses. The Germans have as much inherent propensity to tax evade as the Greeks.

Friday Special edition unknown, now back on Friday!

How to bid on eBay. No, you don't just enter your valuation and relax.

Why men are better at mathematics and engineering.

Eugene Fama: My life in finance.

Open Yale courses.

Interviews of great econometricians.

Mad, mad men.

Beautiful Kokeshi dolls.

Stats and Bayesian econometrics resources

The Foundations of Statistics: A Simulation-based Approach. Nice intro to R too.

A course in Bayesian Econometrics, by Gary Koop.

Friday special special, now on a Monday!

Gary Becker, 2008: We're not headed for a depression. 'World economic growth will recover once we are over the present severe financial difficulties.'

Don't blow your nose when you have a cold.

Very interesting research on dating.

Why are modern scientists so dull?

Did people who knew about secret CIA coups game the market?

In the beginning, a crash in appetite for risk

DeLong, December 2008:

Think of it this way: two years ago we lived in a world in which the wealth of global owners of capital was some $80 trillion — that was the market value of all of their property rights to dividends and contract rights to interest, rent, royalties, options, and bonuses.

In the past two years the wealth that is the global capital stock has fallen in value from $80 trillion to $60 trillion. Savings has not fallen through the floor. We have had little or no bad news about resource constraints, technological opportunities, or political arrangements.

$17 trillion of [the fall] comes by [my] arithmetic from a rise in the risk discount. There has been a massive crash in the risk tolerance of the globe’s investors.

Thus we have an impulse — a $2 trillion increase in the default discount from the problems in the mortgage market — but the thing deserving attention is the extraordinary financial accelerator that amplified $2 trillion in actual on-the-ground losses in terms of mortgage payments that will not be made into an extra $17 trillion of lost value because global investors now want to hold less risky portfolios than they wanted two years ago.

Our models predict that in normal times, with the ability to diversify portfolios that exists today, the risk discount on assets like corporate equities should be around 1% per year. It is more like 5% per year in normal times — and more like 10% per year today. And our models for why the risk discount has taken such a huge upward leap in the past year and a half are little better than simple handwaving and just-so stories. Our current financial crisis remains largely a mystery: a $2 trillion impulse in lost value of securitized mortgages has set in motion a financial accelerator that we do not understand at any deep level but that has led to ten times the total losses in financial wealth of the impulse.

The wisdom of Chris Dillow

The recession does not exist: individuals’ lived experience tells us little about macroeconomic phenomena such as booms and slumps. These are aggregate economic data, not direct sense data to any individuals.

An excellent piece by Chris Dillow on the inability of individuals to directly perceive macroeconomic phenomena. Yes - it's still clearing out my closet season, but hey - the recession's still here so I'll count this one as timely.

In more recent blogging, I like how Chris has something interesting to say, even when commenting on the most boring story in the world:

“Management“ functions rather like witchcraft. It’s a set of rituals which are wrongly supposed to have effects on the outside world.

by datacharmer | Thursday, February 02, 2012
  | 0 comments | | The wisdom of Chris Dillow @bluematterblogtwitter

Schools test often to teach inequality and hierarchy

At school, our kids are rated and ranked far more often than most adults will tolerate, even though this actually slows their learning! It seems that modern schools function in part to help humans overcome their (genetically and culturally) inherited aversions to hierarchy and dominance. Modern workplaces require workers who are far more accepting than are foragers of being told what to do when, and of being explicitly ranked, and our schools prepare kids to accept this more primate-like environment.

The evidence strongly suggests that students learn better when they are not graded and certainly not when they are graded on a curve.

Subjects worked on different tasks and received performance-contingent payments that varied in amount from small to very large relative to their typical levels of pay. With some important exceptions, very high reward levels had a detrimental effect on performance.

This is Robin Hanson of Overcoming Bias making a plausible point, though I would not view this at all as a recent phenomenon. If anything, accepting hierarchy and dominance is probably less important today than it used to be at any point since pre-history.

Do click through, there's more interesting stuff there.

Gelman galore

Clearing Out My Closet season continues, this time with a number of interesting posts from Andrew Gelman:

Uh.. Um..
I [Mark Liberman] took a quick look at demographic variation in the frequency of the filled pauses conventionally written as “uh” and “um”.

Marginal vs marginal
To me, the most interesting bit of terminological confusion is that the word “marginal” has opposite meanings in statistics and economics. In statistics, the margin (as in “marginal distribution”) is the average or, in mathematical terms, the integral. In economics, the margin (as in “marginal cost”) is the change or, in mathematical terms, the derivative.

If you are jumpy, you are less likely to be liberal
In a group of 46 adult participants with strong political beliefs, individuals with measurably lower physical sensitivities to sudden noises and threatening visual images were more likely to support foreign aid, liberal immigration policies, pacifism, and gun control, whereas individuals displaying measurably higher physiological reactions to those same stimuli were more likely to favor defense spending, capital punishment, patriotism, and the Iraq War.

Friday the 13th, unlucky?
A study published on Thursday by the Dutch Centre for Insurance Statistics (CVS) showed that fewer accidents and reports of fire and theft occur when the 13th of the month falls on a Friday than on other Fridays. . . . In the last two years, Dutch insurers received reports of an average 7,800 traffic accidents each Friday, the CVS study said. But the average figure when the 13th fell on a Friday was just 7,500.

Datacharmer recently made a good comment on this:

Apart from avoiding risky behaviour on Friday the 13th because it is deemed unlucky (which might well be happening), you should also consider that Friday the 13th – unlike other Fridays – CAN’T be Christmas or New Year’s (where people get drunk and drive), and it will also be associated with a lower (or higher) probability of falling before a bank holiday weekend (or I guess in the States Independence day, etc).

I guess all I’m saying that it could well be other factors driving this result other than a change in people’s behaviour because Friday the 13th is ‘unlucky’.

How about accidents on Friday the 12th of Friday the 14th? The article only compares Friday the 13th with an average Friday – in fact, it doesn’t even reveal whether the 13th is least accident prone Friday in the book…

by datacharmer | Sunday, January 22, 2012
  | 0 comments | | Gelman galore @bluematterblogtwitter

If you can prioritise well you are probably not very creative

In a 2003 study at Harvard, Dr. Carson and other researchers tested students’ ability to tune out irrelevant information when exposed to a barrage of stimuli. The more creative the students were thought to be, determined by a questionnaire on past achievements, the more trouble they had ignoring the unwanted data. A reduced ability to filter and set priorities, the scientists concluded, could contribute to original thinking.

The NY Times has more.

I'm clearing out my inbox of its 400+ emails, and many of these relate to stuff I wanted to post here - so expect a regular stream of content. Even if a couple years old, good links always stay fresh!


OK, this is geeky - but love it.


by datacharmer | Tuesday, January 17, 2012
  | 0 comments | | ASCII 3D @bluematterblogtwitter