|
PureBytes Links
Trading Reference Links
|
Harley,
Oh, I can't let that pass. The Central Limit Theorem (CLT) says nothing of
the
kind! Balderdash! Yours is a common misunderstanding.
The CLT says that the sum, S, of a large number, n, of independent and
identically distributed random variables having a common mean, m,
and variance, v, will be approximately normal with mean nm and
variance nv.
You can divide the sum, S, by the number of variables, n, and in the limit
as n goes to infinity, the mean of S/n will be m and the variance v.
References:
1. "The Probability Turtoring Book", Carol Ash, IEEE Press
2. "Stock Market Probability", Murphy, Irwin
3. "Introduction to Probability Theory"; Hoel, Prot & Stone; Houghton
Mifflin
this last rather rigorous.
Now, look at that if you will. All the CLT says it that you have a normal
distribution. It says nothing about the probabilities of getting runs of
various
sizes.
If S is normally distributed with zero mean and any variance, the
probability
of S being greater than zero is the same as S being less than zero, i.e,
50%. Run the experiment any number of times and get any number of
consecutive runs where S is greater than zero. Guess what? The odds of
getting S greater than zero on your next run of the experiment will be 50%.
Martin
Happy Trading
----- Original Message -----
From: "Harley Meyer" <meyer@xxxxxxxxxxx>
To: <metastock@xxxxxxxxxxxxx>
Sent: Friday, 14 April, 2000 6:42 AM
Subject: Re: Money Management Stops
> Not that I say much these days, but Mickey is not all that wrong. The
> central limit theorem is a key theory in probability. So using a fair
coin,
> the odds are 50 - 50 of getting a head or a tail. The theory, and I want
to
> underscore the theory, states that as the coin is flipped indefinitely and
> approaches infinity. The result will be a 1 in 2 chance of being heads.
But
> the actual event of flipping the coin can produce subsets that contain a
> long string of tails or heads. If a long string of tails occurs, then the
> central limit theorem would then require that the frequency of heads would
> need to increase. So as you look at the raw data the outcomes {H,T} is
only
> 50 - 50 in the limit. And as one member H or T increases in frequency. The
> other must eventually increase in frequency. The problem is that we don't
> know when.
>
> Not to speak for Mickey, but this is the observation that I understood he
> was conveying.
>
> However, one question I have is. What distribution are you using to build
> your statistics? Poison, Gamma, Normal, etc.?
>
> Harley
>
> ----- Original Message -----
>
>
>
> > Rubbish, Mickey.
> > While you still have a bankroll, buy a roll of pennies and start
flipping
> > :o)
> >
> > Bob
> > -----Original Message-----
> > From: owner-metastock@xxxxxxxxxxxxx
> > [mailto:owner-metastock@xxxxxxxxxxxxx]On Behalf Of Michel Amelinckx
> > Sent: Thursday, April 13, 2000 11:10 AM
> > To: metastock@xxxxxxxxxxxxx
> > Subject: RE: Money Management Stops
> >
> >
> > Sorry but I by this "every time you have a losing trade the odds of your
> > next
> > trade goes up" I meant the PROBABILITY of next trade goes up. You always
> > have 50% odds.
> >
> > > If the success rate is 70%, it's true that there is about a
> > > 99% chance of
> > > having 4 losing trades in a row. However, once you've
> > > already had 3 losing
> > > trades, the chance of the next trade being a success is still
> > > only 70%.
> >
> > No I don't agree, a system 70% prof. and having 3 consecutive losing
> > trades - the probability that the next trade will be a successful is
99%.
> > If you have a system that is 55% correct of the time. If you wait for 4
> > consecutive losers you have almost an 95% probability that the next
trade
> > will be successful.
> >
> > Same thing with roulette, they hate me in the casinos because if I play
> > roulette I play on red and black. I wait till red (black)past 3 or 4
> times
> > in a row and then play on the other colour. And the longer you wait,
like
> 5
> > or 6 times on red (black) the higher the probability the next will be
the
> > other colour. Although 6 times the same colour is very rare. And this
they
> > don't like in casinos.
> >
> > Greetings
> >
> > Mickey
> >
> >
> > > -----Original Message-----
> > > From: owner-metastock@xxxxxxxxxxxxx
> > > [mailto:owner-metastock@xxxxxxxxxxxxx]On Behalf Of Randy Harmelink
> > > Sent: donderdag 13 april 2000 17:24
> > > To: metastock@xxxxxxxxxxxxx
> > > Subject: Re: Money Management Stops
> > >
> > >
> > > You can't use statistics that way. An increase in
> > > probability only occurs
> > > if the events are dependent.
> > >
> > > For example, suppose you are trying to draw an ace of spades
> > > from a deck of
> > > cards. If you continue to draw and discard cards that aren't
> > > the ace of
> > > spaces, your probability of drawing the ace of spades
> > > increases on each
> > > draw. But if you put each drawn card back into the deck and
> > > reshuffle, your
> > > probability of drawing the ace of spades on a given draw will
> > > never change,
> > > no matter how many times you fail to draw it.
> > >
> > > If the success rate is 70%, it's true that there is about a
> > > 99% chance of
> > > having 4 losing trades in a row. However, once you've
> > > already had 3 losing
> > > trades, the chance of the next trade being a success is still
> > > only 70%.
> > >
> > > Otherwise, it would be easy to develop a system to beat a
> > > roulette wheel.
> > > <G>
> > >
> > > ----- Original Message -----
> > > From: Michel Amelinckx <Michel.Amelinckx@xxxxxxxxxx>
> > > To: <metastock@xxxxxxxxxxxxx>
> > > Sent: Thursday, April 13, 2000 7:58 AM
> > > Subject: RE: Money Management Stops
> > >
> > >
> > > > And because you have such a great number of profitability,
> > > did you know
> > > that
> > > > every time you have a losing trade the odds of your next
> > > trade goes up.
> > > > 70% prof - after 2 consec losing trades - probability next
> > > trade will be a
> > > > winner is 91%
> > > > 70% prof - after 3 consec losing trades - probability next
> > > trade will be a
> > > > winner is 97%
> > > > 70% prof - after 4 consec losing trades - probability next
> > > trade will be a
> > > > winner is 99%
> > >
> > >
> > >
> >
>
>
>
|