From: John Conover <john@email.johncon.com>
Subject: Re: Fwd: [RT] Buffett Forecasts Eight Bad Years
Date: 28 Aug 2001 11:51:36 -0000
BTW, that is the way:
http://www.johncon.com/ntropix/FAQs.html#calculation
was done-which turned out to be prophetic.
The rules of the game were that I could use only daily, monthly, and
yearly, graphs, (not data,) of the US exchange indices, and could use
no computer programs, calculators, strait edges, etc.
If you look at it, it was not that difficult. The last loosing year
for the US indices was 1993. By 1998, (when that series of
prognostications started,) which was about 5 years into the "bubble",
the chances of it continuing at least one more year would be
approximately 1 / sqrt (5), or call it about 50%. So, forecasting a
downturn was my best choice.
Note that since equity indices are fractal, (self similar and affine,)
I "extrapolated" the data to make a daily, weekly, monthly, yearly,
and a decade prognostication.
John
BTW, since I could not use a calculator-and had to take square roots
in my head-I used "Continued Fraction Expansion of the Golden Mean,"
which was known to the ancient Pythagoreans.
It is an iteration:
x = (x + 2 / x ) / 2
n + 1 n n
which after a reasonable guess converges to 4 decimal places in 4
iterations-if you can remember 4 decimal places to carry forward to
the next iteration.
The technique is historical in mathematics-it was used to prove the
incommensurability of the square root of 2, (e.g., sqrt (2) != p / q
for any integers, p and q-upsetting the prevailing wisdom of the
Pythagoreans,) which lead to the Diophantine equations, (and Fermat's
last theorem-which was only resolved in the last decade,) which lead
to Cantor's numbering scheme, (proving that there are more real
numbers than rational,) which was a mapping exploited by Godel in 1928
to show that in any complex system, knowledge of the system must be
inconsistent or incomplete, a concept which was further expanded by
Chaitin in 1992 to prove information-theoretic incompleteness, (e.g.,
there are more undecidable propositions than not,) which is why things
human are fractal.
Depending on who is telling the story, of course.
John Conover writes:
> How about 1 / sqrt (8) = 35.35%, or about 1 in 3.
>
> So, if you wanted to wager against Mr. Buffett's analysis, you would
> bet that he has a 65.65% chance of being wrong.
>
> And, you would bet 2 * 0.6565 - 1 = 29.29% of what you could afford to
> loose on the wager.
>
> John
>
> Jeff Haferman writes:
> >
> > Hmmm, I don't believe an entropic analysis would give a very
> > high probability to 8 years of stagnation....
> >
> >
> > >From Businessweek (9/3/2001)
> > ---------------------------
> > THE ECONOMY
> > Buffett Forecasts Eight Bad Years
> >
> > If you're looking for a rosy economic forecast, don't knock on Warren
> > Buffett's door. Seems the Berkshire Hathaway chairman and King of All
> > Value Investors has been telling the executives he meets with to
> > brace themselves for a long slowdown. Not only is there no turnaround
> > in sight this quarter or even this year, according to Buffett, but
> > those who've met with him say that he is predicting eight years of
> > economic stagnation. Buffett attributes the standstill to a "hangover
> > effect" from the excesses of the late 1990s, says one private equity
> > investor who has heard the Sage of Omaha's reasoning.
> >
> > Buffett generally doesn't speak to the press, and questioning his
> > headquarters about his economic forecast proved to be no exception.
> > But if the 70-year-old business guru is right--as he so often has
> > been in the past--his strategy of rational investing for the long
> > term is going to look even smarter down the road.
--
John Conover, john@email.johncon.com, http://www.johncon.com/