From[Bill Williams 31 May 2004 1:20 AM CST]
[From Rick Marken (2004.05.31. 2200)]
Bill Williams (31 May 2004) --
It's clear from the "models" you've posted that you don't know
how to write a computer model
How, would you know? You couldn't figure out how the Lattice model
worked. Initially you said it didn't have a control loop. Basically
you have been looking at stuff you don't understand.
The Lattice model definitely had no control loop in it.
You understand even less than I thought you did-- which wasn't much.
The lattice points -- which I suspect were the variables you
thought were being controlled -- were not controlled. The
program was simply an overly complicated array plotting routine.
Rick, you can't know how pleased I am to have this report from you.
But, can I ask where do you suppose the array that you think is
being plotted is stored?
This assumes that Bill Powers' criticism was correct. It is easy enough
to show that Bill Powers' understanding of economic modeling is not at
all on a part with his modeling of other simpler phenomena.
This is not a matter of understanding economic modeling. It's a matter
of understanding how to implement models in programs. Your programs do
not properly implemented control system models.
Well, since you have demonstrated that you can't recognize a control loop
when you see one (such as in the Lattic model) there is no reason to take
your claim to heart. And, Bill's criticisms of my economic models have been focused on those models and constructs which have been included in Keynesian
models. I best I can understand it, and as Bill Powers has admitted, Bill Powers doesn't understand the Keynesian system. I don't think you've ever claimed to do so. It really did perturb me initially the way Bill reacted to the Keynesian models. I wondered could I have make that serious an error? But, then I realized that what Bill Powers was saying about Keynes was precisely what he was saying about me. And, then I felt better, I even felt good. You can't know how nice it is to be compared to Keynes-- even if both of us, Keynes and I are being called "garbage."
Rick I think it is time for you to send me a letter acknowledging that
you without my approval altered the Giffen paper that was published in
the American Behavioral Science issue. While you may not "know
everything" you think that you know more than I do. You even think you
know more about the Giffen effect than I do. This mistaken notion of
yours that you understand economics has created problems. When you
decided to improve the Giffen paper you introduced errors that I
didn't make. Those errors are your responsiblity. But, anyone reading
that issue would think that they are my errors. So, send me a letter
211 Haag Hall
University Missouri-Kansas City
5100 Rockhill Road
Kansas City, MO 64110
I'll be happy to. I'll write it up as soon as I can. I'll mail it
before the end of the week. But in order to write it up properly I
would appreciate it if you could send me a description of the errors
that I introduced into the paper. Just copy the sentences in the
article that are in error (or, if it's easier, just list the line(s)
and page numbers of each error) and explain why they are errors.
It isn't that hard to explain. I remember what you did. Normal demand functions are downward sloping to the right- that is as the price falls
there is an expectation that more of the product will be demanded. The Giffen effect violates this expectation-- the demand for the inferior good decreases when the price decreases and increases when it increases. The demand for the superior good in the Giffen case increases when income increases up to the point at which the caloric requirement can be supplied entirely through the consumption of the superior good. After that when increases still further the demand for the superior good is entirely inelastic-- that is there is no further increase in the consumption of the superior good as income increases. You said as, I remember it, that the demand for the superior good would be normal beyond the point at which the budget reached a point at which the caloric requirement could be met by the superior good. However, this is not the case-- even in respect to the superior good, the consumer is controlling for (familiar concept? ) the consumption of a reference level of calories. There isn't any reason at all for the consumer to consume more calories just because income increases. The difference between asserting that the demand above the Giffen region is normal and entirely inelastic with respect to changes in income involves the question of whether or not the superior good behaves in part in the way that neo-classical economic theory predicts, or whether the superior good having behaved according to orthodox prediction in the giffen space then also behaves normally outside the giffen space. Well, interesting outside the giffen space the superior good then behaves abnormally outside the Giffen space. And, this is important because studies of consumer behavior routinely report that consumer demand is much less elastic than the neo-classical conception of consumer behavior would lead one to suspect.
So, you rewrote the paper by changing a valid prediction that the model makes,
a prediction that is contrary to orthodox theory, into a prediction that is inconsistent with the model, and also inconsistent with the finding that consumer behaivor is much less elastic than the orthodoxy's Economic Man would lead one to think.
Would you like for me to try to explain to you where the control loops are
in the Lattice program.
From: Control Systems Group Network (CSGnet) on behalf of Rick Marken
Sent: Mon 5/31/2004 11:58 PM
Subject: Re: FW: Social Reality