On Smoking Cannabis, Cavemen, Giants and Predicting Stock Markets

Once upon a time, in a land far far away, lived a hairy caveman. After a long hard day of hunting mammoths and avoiding saber-tooth tigers, he scratched his nose and it started raining. He didn't think much of it at first and ran back to his cave into the arms of his loving cavewoman. The next day he scratched his nose again and voila! It rained again. The caveman came to an important conclusion: "if" he scratches his nose, "then" it rains. Sometimes it worked, sometimes it didn't, but at least he had a way to cope with the randomness of nature and had assurance that he had some control over the arbitrary events of his life. He spent the rest of his days devising an intricate system to optimize the results of his nose-scratching activity. It was the birth of the "science" of prediction.

Randomness is scary, and unfortunately we all have a little bit of caveman in us. We would like to think that "if" something happens "then" something else will happen. We are so afraid of randomness that we are willing to pay exorbitant fees to ivy-league educated Wall Street suits to tell us "if xyz happens then you should buy abc." Would you pay a caveman to tell you that "if" you scratched your nose "then" it will rain? Eric Hoffer, an American social philosopher, said:

Creativity is the ability to introduce order into the randomness of nature.

As I've mentioned in THE PANELIST, statistics, which aim to introduce order to randomness, is often an act of creativity and not always one of science. Given the amount of statistical "facts" and their wide spread exploitation by the mass media, it surely pays to be at least aware of the dangers.

Whenever you are faced with a statistical claim, remember that CORRELATION DOES NOT IMPLY CAUSATION! Sometimes changes in two correlated quantities are both the result of a third factor. A few examples might be useful:

1. Children with bigger feet spell better. Does this mean "if" a child has bigger feet "then" he/she will spell better? No! children with bigger feet spell better because they're older, not because they have bigger feet.

2. People who sleep with their shoes on tend to wake up with a headache. Does this mean "if" I go to sleep with my shoes on "then" I will wake up with a headache? Of course not! If you fall asleep with your shoes on, chances are you had one too many drinks and passed out. Happy hangover!

3. People who use cannabis have a higher prevalence of psychiatric disorders. Does that mean "if" I smoke cannabis "then" I will go crazy? Statistically this isn't correct. All the relationship tells us is that people who use cannabis are more likely to develop a psychiatric disorder, and there could be a third factor influencing the relationship.

With the above examples in mind, let's move on to one of the stock market's quirkiest - but most accurate - "indicators." The Super Bowl Predictor of stocks has called the right direction of the Dow Jones Industrial Average for the year following 33 of the 41 bowls, or an astonishing 80.5% success rate. Blogger David Gaffen explains:

In short, whenever an 'original' National Football League team wins the big game, the market rises; but it falls when the winner is a team like the New England Patriots that joined the NFL after the league's merger with the American Football League in 1970.

He adds:

Last year, the indicator worked yet again. The Indianapolis Colts, an original NFL team based on their roots as the Baltimore Colts, won the game -- and the stock market had a winning year.

In other words, "if" the Giants had to win Sunday's Superbowl, "then" stocks will go up over the next year (go Giants!)

Nothing I have written so far is new. Given the vast amount of literature dedicated to these ideas, I find it puzzling that the Wall Street choir always sings "if" the Fed cuts interest rates "then" the stock market will go up. But the beat goes on, and the choir has a lot of cheering to do after the Fed cut interest rates by 1.25% a few days ago.

It's NOT a certainty that lower interest rates will automatically lead to stock market gains. "Flow follows performance," says Charles Biderman of Trimtabs.com. "Also has, always will. If stocks go down, money will leave equities, even if interest rate is zero. Japanese investors ignored Japanese stocks for a decade even as Japanese interest rates hovered around 1%." With the U.S. mortgage market in tatters, will there be enough pricing power to push stock markets higher? For a more thorough analysis of this idea, click here.

Disclaimer: The caveman metaphor comes from Fooled by Randomness, a great book written by Nassim Nicholas Taleb. Read it!

Comments

[mbaty]

If a Democrat takes office Then our problems will be solved.

[realitytrumpsbull]

I closed my investment account, and the people holding my 'retirement' money haven't called back. Hmmm....hm­mm....hmmm­....I think that whole business is a fraud, too.

[Dandy12]

This is to say absurdity rules... The entire society in all respects has no sense of balance and proportion. This administration has exacerbated the problem in all respects!

[Sciguy]

The caveman's behavior is called "superstitious behavior" in the operant behavior biz. You'd think that by now, we'd have collectively outgrown such behavior - but not a chance. I wonder if it has something to do with how we learned and survived when we were cavepeople. For example, if something seemed to work, like throwing rocks at predators (or at live food), it helped us survive. We kept doing it, for everything, just in case (not consciously "just in case" but hardwired "just in case").

As for the statistics - as attributed to Benjamin Disraeli: "There are three kinds of lies: lies, damned lies, and statistics." Such is life.

[jimmyaj]

Anyone so credulous as to believe in prognostication by superstitious methods will be at a disadvantage because he will be so certain of the outcome he will not be observing actual events that might bear on the results, changing the prediction. The churches are full of such people.

[WIpatriot]

Can we do reverse causation?

Since the stock market is going to go down this year, the New England Patriots will win the Super Bowl on Sunday.

[DemandTruth]

I believe Colbert and Conan and Stewart were recently applying this bit of causation regarding Huckabee! A right treat!

ps. doesn't anyone consider that people who are more predisposed to psychological disorders or disturbance or whatever, be it because of nature (like inherited addictive tendencies, etc) or because of nurture (problems in the home, abuse, etc) are MORE likely to self-medicate, and one thing with which you can medicate is cannbis? Did the brilliant scientists think of that? I certainly know MANY people who self-medicate for anxiety or depression with pot. And to me, a natural herb w/ no real harmful side effects sounds better than ANY prescription drug - psychological or otherwise with all their crazy myriad side effects!

[LeftRight]

Don't forget the other one. "If" you cut taxes "then" everything will be right with the economy.

[cynara]

Oooh! OOooh! Let me try!

Here's another one for you: Latinos voted for Hillary Clinton in Nevada, therefore they didn't vote for Obama, and since Obama is black - Latinos won't vote for a black man. Okay, so its not a perfect enough analogy. But its close enough to stupid to work for me, considering that the political analyst don't have another piece of evidence to go on besides this one sequence of events.

And yet, they proclaimed it far and wide, without even stopping to consider the possible third factor influencing the relationship - a large percentage of Latinos (a much larger percentage than the general population) are new voters, new voters tend to be less informed, and even more so if their primary language isn't English (as is true with decent percentage of Latinos) and if they are blue collar burned out workers who don't have time to follow politics (also more true in the Latino population than the general population). People with less candidate information will tend to vote more on name recognition than people with more information. So, lets see, who would have a higher name recognition "Obama", "Edwards", or "Clinton"?

Nahh, that couldn't be it. Latinos are racist against blacks, the pundits and pollsters told me so...