Random Walk Hypothesis / Word Usage
[In the "authority" page of the "Children of Narcissus" article]: "A stock that has doubled in value in 12 months is way ahead of the market average and isn't statistically likely to continue that pattern of growth"
That assertion runs counter to the random walk hypothesis, does it not?
No, for these reasons:
- Random walks follow a stochastic pattern (one determined by randomness and probability). While a small move is likely, a large move is less likely because such a move is normally (but isn't required to be) composed of a series of small moves. This doesn't mean there are no large moves, it means they are less probable.
- Equity markets respond to human emotion and fundamental issues like P/E ratios, not just stochastic moves, about which more below.
My understanding of random walk is that valuations have no momentum,
It's not a question of momentum, it's a question of the relationship between step size and probability, something that looks like momentum to a first approximation.
so random stocks A and B are equally likely to double or exceed the market average in the coming year, regardless of whether they did either of these things in the trailing year.
That is a correct statement of the Gambler's Fallacy, which is true — the future of a random process owes no allegiance to its past. But the equities market follows different probabilities, because of the role of human psychology and market fundamentals.
Remember there is a cause-effect relationship between any particular stock and the overall market averages — it is difficult for a stock to escape the expectation that it should follow the market up and down. And because individual companies are normally components of larger interdependent assemblages, this tracking effect is reasonable.
Also, for any particular stock, a sudden infusion of capital must be followed by growth in the company's profits. If this doesn't happen, if an infusion of capital is not followed by an increase in the company's profits, further price increases will be stymied by the resulting unfavorable P/E ratio. So a sudden increase in a stock's price is less likely to be duplicated unless and until the company responds to the first increase, something that is by no means guaranteed.
Remember about the random walk hypothesis that, unlike a pair of dice or a roulette wheel, some market players pay attention to the relationship between a stock's present price and the company's present earnings. This attention to fundamentals tends to tamp down wild swings in price, especially repeated ones.
For example, the 2000 tech bubble burst because people finally realized the tech companies would never redeem the faith expressed by the absurd P/E ratios of that era, so investors pulled out all at once.
The same thing just happened again — investors finally realized that banks' published P/E ratios weren't truthful. The P/E ratios didn't take into account all the bad commercial paper that banks had managed to keep off their books, so once investors realized this, they pulled out.
Banks had been using accounting tricks to hide an unsustainable level of debt, to get around banking rules that require a particular ratio of liquidity to debt. But even though the debt was off the banks' books, the consequences of default were not.
These are factors not taken into account in the random walk hypothesis, which basically argues that the market is a stochastic process, without emotion or attachment. The is true only to a first approximation.
I submit this with some trepidation because once I hastily pointed out to you that you used "malice aforethought" when you meant "malice of forethought". Of course I hear "malice aforethought" regularly now. I asked my English-as-a-second-language wife if she knew it and of course she did. Nature has a fondness for serving me with these cock whippings.
- Apparently the "malice aforethought" construction has been with us since at least 1581:
Etymology of "afore"
It is also regularly used in a legal context:
Malice Aforethought (Wikipedia)
- You know, when I deal with a language issue, I can be as cowardly as anyone, but I have a method. First, I realize that language is not a top-down system (dictionaries don't tell us how to speak), it is definitely bottom-up (we tell dictionaries what to contain). Based on that, when in doubt I Google for a word or phrase to see how real people are speaking and writing — and the form with the biggest number of hits is the winner.
Example:
Compare this:
http://www.google.com/search?q=hi-tech
To this:
http://www.google.com/search?q=high-tech
As expected, "high-tech" prevails over "hi-tech", but not by very much:
"high-tech: 89 million
"hi-tech": 45 million
- Another example with a different solution. Someone once wrote me to suggest that spell checkers could be much more useful than they are. She gave as an example the English rule that says "I before E except after C". So I decided to check the rule — Linux systems have a rather large spell-check dictionary installed by default, so I decided to use it to make a statistical determination (this is a case where Googling is less useful).
Since you are running Linux, [I refer to my original correspondent, but why aren't you running Linux?] you can repeat this experiment for yourself. Open a shell session and type:
$ grep "cie" /usr/share/dict/words | wc -l
result: 863
(the '$' at the left is just to remind you that it's a shell session, you don't type it)
Now the other form:
$ grep "cei" /usr/share/dict/words | wc -l
result: 300
Okay. This says that, in a large English spelling dictionary, "cie" is more than twice as common as "cei". But this result contradicts the rule:
(source: http://en.wikipedia.org/wiki/I_before_E_except_after_C)
"in words where i and e fall together, the order is ie, except directly following c, when it is ei."
So much for spelling rules.
- One more example. Did you know that "literally" can now mean "figuratively"? When I discovered this I was astonished.
http://www.merriam-webster.com/dictionary/literally
Definition 2 : in effect : virtually
This makes no sense, and by so doing proves that language is a bottom-up system. More on literally/figuratively:
The Word We Love To Hate (Slate)
I admit scoffing when you postulated that religion might be a side-effect of brain evolution. Soon thereafter I read http://ffrf.org/fttoday/2003/april/index.php?ft=sapolsky which vindicates your position.
Either that or Sapolsky got the idea from me, in which case it could be an example of two great minds occupying the same gutter.
It's a fascinating and funny read, and describes how OCD and schizophrenia can be what Richard Dawkins calls an evolutionarily stable strategy (ESS), of which Tit for Tat is a model of sorts.
There's plenty of evidence for "Tit for tat" as a survival strategy, much more than for a stable narcissism/enabler ratio in groups. It's possible this is only true because we haven't figured out how to detect narcissism in amoebas.
Similarly narcissism might be a feature, not a bug, from an evolutionary point of view. The fact that most people aren't narcissists, and expect others to be like them, might in of itself guarantee that a certain portion of the population is genetically programmed to be narcissistic.
Or in group dynamics there might be a requirement for a certain number of narcissists as well as a requirement for enablers, in a particular stable ratio. But one thing is certain — you can't have too many narcissists, because they are parasites, and parasites can't survive except within a larger population of hosts.
See http://biomed.brown.edu/Courses/BIO48/16.Evol.Behavior.HTML and http://eurekalert.org/pub_releases/2008-10/osu-npm100708.php (I think I got the latter link from you, but nevertheless I can't find it on your site).
Yep, you did. It's in the references section of the "Narcissism" page of "Children of Narcissus". I see I need to consolidate my reference lists.
... here's the nickel version. Suppose there is a society with no narcissists. In that case, a narcissist is guaranteed to spring up because his unnatural confidence will make him a leader and afford him better opportunities than his peers. If narcissists become too numerous, the society catches on and begins penalizing narcissists, whose numbers then decline. All else being equal the number of narcissists will settle on a fixed proportion.
Honest, I didn't read ahead in your message when I made this same point above.
But all light banter aside, my article is more about the dangerous variety of narcissist, the kind that psychologists label "malignant." These people aren't part of a natural balance between enablers and narcissists — they are really parasitic. They carve a place for themselves out of the flesh of those who happen to be standing too close.
Thanks for writing.
My Narcissism
As the second gulf war turned south and we didn't find any weapons of mass destruction; we only succeeded in weakening our country, doing Bin laden's work for him; as the dead and maimed U.S. soldiers piled up, I turned against George Bush. At first, being young and having not lived through the Vietnam war, I was shocked and I pitied Bush for how badly the war had turned, but slowly I began to blame him for the causalities and costs of the war. I wondered how he could live with himself, and then it dawned on me, that I shared his blame, because it was me, in the lead up to the war, who went to a rally in my downtown, holding a sign: "Free the Iraqi People." I guess this is the narcissism you speak of.
No, actually the narcissism I speak of is the sort that never acknowledges error — the sort of person who, when confronted by evidence of his own missteps, claims he was misinformed and holds himself blameless. The sort of person who, if given a chance, would do it the same way again.
What you describe is called "growing up." Narcissists never grow up.