Markets Re-visited

CSGnet Archive CSG2004
1

[From Bruce Gregory (2004.0828.1511)]

"Suppose you can simulate on your computer an artificial stock market.
Based on your ideas about how one piece of the economy connects to
another, you build an elaborate econometric model. It inputs data you
give it about the weather, inflation, economic growth, industry
specialization, and the companies being traded; and it calculates what
its algorithms tell it is the optimal "fundamental" value of the
company's stock. To that value, it adds millions of small random
changes--perhaps reflecting fictitious news events, or fickle investor
preferences, some taken separately, some added together. So, what kind
of variability generator will you use? If mild, the resulting price
charts will vary within a certain well-defined range; their trace will
be the product of many small computer-generated events. Very different
is wild variability, even though it can be "tuned down" to be less
extreme than Cauchy's. Wild price charts will be a hair-raising record,
mixing small movements with very, very large dislocations, many
computer-generated news items with a few cataclysmic bulletins, many
small transactions with large institutional block trades--all, a mix of
the small and routine with the large and rare. In such a wild world, an
imaginary investor participating in this economic simulation could be
wiped out overnight."

_The (Mis)Behavior of Markets: A Fractal View of Risk, Ruin, and
Reward_ by Benoit Mandelbrot and Richard L. Hudson.

Bruce Gregory

"Great Doubt: great awakening. Little Doubt: little awakening. No
Doubt: no awakening."

2

[From Bill Powers (2004.08.29.0759 MDT)]

Bruce Gregory (2004.0828.1511)--

"Suppose you can simulate on your computer an artificial stock market.
Based on your ideas about how one piece of the economy connects to
another, you build an elaborate econometric model. It inputs data you
give it about the weather, inflation, economic growth, industry
specialization, and the companies being traded; and it calculates what
its algorithms tell it is the optimal "fundamental" value of the
company's stock. To that value, it adds millions of small random
changes--perhaps reflecting fictitious news events, or fickle investor
preferences, some taken separately, some added together. So, what kind
of variability generator will you use? If mild, the resulting price
charts will vary within a certain well-defined range; their trace will
be the product of many small computer-generated events. Very different
is wild variability, even though it can be "tuned down" to be less
extreme than Cauchy's. Wild price charts will be a hair-raising record,
mixing small movements with very, very large dislocations, many
computer-generated news items with a few cataclysmic bulletins, many
small transactions with large institutional block trades--all, a mix of
the small and routine with the large and rare. In such a wild world, an
imaginary investor participating in this economic simulation could be
wiped out overnight."

This is the conclusion one is led to by a model without purposive agents in
it, or a model in which they are overlooked. You could say very similar
things about human behavior in general, couldn't you? The stock market is
only one example of it. If you look only at behavior, you see just as much
variability as there is in all the disturbances that are going on. But if
you look at controlled variables, suddenly exquisite regularities appear.

Any economic model that is not organized around the purposive agents who
run it is going to be hard put to reveal any regularities.

Best,

Bill P.

3

[From Bruce Gregory (2004.08.29.1054)]

Bill Powers (2004.08.29.0759 MDT)

Any economic model that is not organized around the purposive agents
who
run it is going to be hard put to reveal any regularities.

I suspect there is broad agreement as to the motives of the purposive
agents who buy and sell stocks. They want to make money and avoid
taking losses. That said, are we ready to propose a model of the stock
market that reveals the underlying regularities?

Bruce Gregory

"Great Doubt: great awakening. Little Doubt: little awakening. No
Doubt: no awakening."

4

[From Bruce Gregory (2004.0829.1117)]

Let me be a bit more specific. The agents that buy and sell stocks want
to perceive that the price of the stocks they have already bought is
increasing. Further, they do not want to perceive that they are selling
a stock for less than they paid for it. Is that enough?

Bruce Gregory

"Great Doubt: great awakening. Little Doubt: little awakening. No
Doubt: no awakening."

5

From [Marc Abrams (2004.08.29.1218)]

[From Bruce Gregory (2004.0829.1117)]

Let me be a bit more specific. The agents that buy and sell stocks want
to perceive that the price of the stocks they have already bought is
increasing. Further, they do not want to perceive that they are selling
a stock for less than they paid for it. Is that enough?

You will never be able to 'predict' the stock market because the 'prices'
are an _emergent_ property of human behavior. No one individual is
responsible for any one price yet each price is due in part to the
activities of each individual and the economic processes involved.
The stock market has as much to do with the 'economy' as 'income' has to do
with 'poverty'. That is, the stock market reflects how people _feel_ about
the economy. Google is valued at more than GM. Do you believe that this
reflects the actual capital held and profits made by each? Please. The stock
market 'crash' in 1999-2000 was due to the 'market' (that is, people)
realizing that these dotcom's were worthless and their valuations
meaningless.

I wish Bill Powers would realize how close Bill Williams is to helping PCT
move along. Wise up Bill P and get yourself a book on microeconomic theory.
Or _Basic Economics_ by Thomas Sowell as Bill Williams and I both
recommended a few weeks ago.

Your stubbornness is _not_ a virtue. Get over it, admit for once in your
life that you've made a mistake and move on. You are, after all human.

Your refusal to communicate with me or Bill Williams is _not_ to the benefit
of your theory or the rest of CSGnet. Mary, I wish you could penetrate that
very thick skull. He is so damn close, yet light years away.

I have a strategy that I think will work, but I will not reveal it on CSGnet
until Bill P is willing to communicate with me.

Marc

6

[From Bill Powers (2004.08.29.1127 MDT)]

Bruce Gregory (2004.08.29.1054)]

Bill Powers (2004.08.29.0759 MDT)

Any economic model that is not organized around the purposive agents
who
run it is going to be hard put to reveal any regularities.

I suspect there is broad agreement as to the motives of the purposive
agents who buy and sell stocks. They want to make money and avoid
taking losses. That said, are we ready to propose a model of the stock
market that reveals the underlying regularities?

I was really referring to economic models in general. The stock market
redistributes money in about the same way they do it in Las Vegas, but I'm
not sure what its real functions in the economy are. As you say, it's
organized around people's goals and it has certain established procedures
for enabling them to pursue those goals.

The kinds of regularities that a model describes or reveals are stated
conditionally: if this happens, that results. For a control system, we say
that if there is a disturbance, an action will take place that opposes its
effects on a controlled quantity. But large systems have many complex
interactions built into them, few of which are comprehensible without a
working model to show how things work.

Of course what most people want from a model of the stock market is a
secret that will make them rich. That's not quite the same thing as
understanding how the system works. Sometimes understanding how the system
works shows you that there's no way to get rich without cheating, as in a
casino.

Think of a model of human behavior. Does it explain what a person will do?
You have to know how a person is organized inside to do that -- that is,
the specific things the person wants at every level. And you have to know
what is happening around the person, too. So predicting (and controlling)
behavior can't really be the point of PCT.

In economics, a good model would let you answer what-if questions, like
"What if a large fraction of producers decided to cut costs by laying off
their workers?" That might lead to some level of prediction, but it would
be concerned with large-scale effects, not individual cases.

Best,

Bill P.

7

From [Marc Abrams (2004.08.29.1437)]

Even though I am currently having a conversation with myself...

[From Bill Powers (2004.08.29.1127 MDT)]

Sometimes understanding how the system
works shows you that there's no way to get rich without cheating, as in a
casino.

Yes, understanding the 'system' might be a first good step in understanding
what you might be able to say about the system. There are plenty of ways to
make _lots_ of money in the stock market without 'cheating', just as there
are an _equal_ number of ways of _losing_ money in the stock market.

Think of a model of human behavior. Does it explain what a person will do?

It should, not _specifically_, but it should provide an understanding of
_why_ we behave as we do. You have discovered the _mechanism_ that 'drives'
it, not all the factors that are involved in it. There is more to a watch
than the gears. You can't seem to understand that people want to tell time,
not watch the gears move.

You have to know how a person is organized inside to do that -- that is,
the specific things the person wants at every level. And you have to know
what is happening around the person, too. So predicting (and controlling)
behavior can't really be the point of PCT.

What is the 'point' of PCT if not to understand why we do what we do? What
is the payoff in learning it? Can anyone besides Bill tell me? Does anyone
care?

Why do you folks want to learn PCT? I'm interested in knowing why folks on
CSGnet think PCT is important.

In economics, a good model would let you answer what-if questions, like
"What if a large fraction of producers decided to cut costs by laying off
their workers?" That might lead to some level of prediction, but it would
be concerned with large-scale effects, not individual cases.

You don't know what you are talking about. It might seem like a harsh
comment but you really have no clue. You seem to be so wrapped up in an
ideology, that you can't see the forest for the trees. With 'questions' like
these, how do you expect _any_ economist to take you seriously?

What a shame and what a waste. _FOR ALL CONCERNED_.

Marc or CRWF