Combining multiple systems

Quote from phattails:

So, let me get this straight... you have key statistics that you use for the sub-categories that ulimately tell you how to allocate your money. And you have categorically different systems that relate to each other under a certain condition, and you're trying to figure out how to measure and/or categorize them. Is this correct? Do you notice this sort of thing with other different systems or just this example?

Why not categorize your functions by an attribute of times
(shocks in security types,market volatility, economic conditions, interest rates,correlation to portfolio, contango/backwardation, yieldcurve,etc.) ? If you're trading off of these signals without an explanation, then you could create variables for the relationships.


Out of curiosity... how do you compare the categories-- what do you do after you have your grand measure?

1. Key statistics don't tell me shit. It's a measure. cold, hot, hotter, hottest... etc. etc.

2. I'm always testing for something better for anything.

3. Different systems? Is this semantical stuff... systems meaning portfolio models? For discussion purposes, let's keep systems as systems and portfolio systems as... Portfolio Models????

4. I've got others. Running multiple "portfolio models" may be complicating for a few but you just need to get used to it. I've over-simplified what I wrote. The key to monitoring multiple "portfolio models" consists 2 things. You need to program both the logic (obviously) and the presentation (An easy to use Windows GUI) so that it's easy to manage.

5. Categorizing by functions are too parametric. The functions you mention are what changes in the market, not something to be used as a category.

Seriously, don't you notice that what you mention is something that has to be dealt by the "trading system"? I get a feeling you're relating your questions based on single system development process. Though similar, the approach is different. Interesting how you mention "these signals". Seriously, don't you notice that what you mention is something that has to be dealt by the system?

Also, what you mention does not include in my list of "tendencies".

...
 
One of the reasons that makes Parrodo's paradox so attractive is the Monty Hall problem.

"The Monty Hall problem is a probability puzzle loosely based on the American television game show Let's Make a Deal. The name comes from the show's host, Monty Hall. The problem is also called the Monty Hall paradox, as it is a veridical paradox in that the solution is counterintuitive.
A well-known statement of the problem was published in Parade magazine:
Suppose you're on a game show, and you're given the choice of three doors: Behind one door is a car; behind the others, goats. You pick a door, say No. 1, and the host, who knows what's behind the doors, opens another door, say No. 3, which has a goat. He then says to you, "Do you want to pick door No. 2?" Is it to your advantage to switch your choice?
Because there is no way for the player to know which of the two unopened doors is the winning door, most people assume that each door has an equal probability and conclude that switching does not matter. In fact, in the usual interpretation of the problem the player should switch—doing so doubles the probability of winning the car from 1/3 to 2/3. Switching is only not advantageous if the player initially chooses the winning door, which happens with probability 1/3. With probability 2/3, the player initially chooses one of two losing doors; when the other losing door is revealed, switching yields the winning door with certainty. The total probability of winning when switching is thus 2/3.
When the problem and the solution appeared in Parade, approximately 10,000 readers, including nearly 1,000 with Ph.D.s, wrote to the magazine claiming the published solution was wrong. Some of the controversy was because the Parade version of the problem is technically ambiguous since it leaves certain aspects of the host's behavior unstated, for example whether the host must open a door and must make the offer to switch. Variants of the problem involving these and other assumptions have been published in the mathematical literature.
The standard Monty Hall problem is mathematically equivalent to the earlier Three Prisoners problem and both are related to the much older Bertrand's box paradox. These and other problems involving unequal distributions of probability are notoriously difficult for people to solve correctly, and have led to numerous psychological studies. Even when given a completely unambiguous statement of the Monty Hall problem, explanations, simulations, and formal mathematical proofs, many people still meet the correct answer with disbelief."

This when used with the original question about multiple systems might lead to a very unusual outcome. I have always treated the Parrondo's paradox in combination with the Monty Hall problem.
 
hmm. the best explanation i ever heard is this. when
you have three doors in the beginning and you pick
one, then the likelihood of having the car is 1/3 and
the likelihood of having not (=it is in one of the two
others) is 2/3. what happens when the moderator
opens one of the two that you did not pick is that the
initial 2/3 is still 2/3, but that probability is now, by the
additional knowledge of the moderator concentrated on
just one door. so you have one with 1/3 chance to win,
and one with 2/3 chance to win.

if you want to claim that a trading problem has relevant
similarity to the 3 door story, there need to be certain
relationships between the systems in place. i struggle
to find the link to trading. it is very clear that the moderator
is the critical element here. his act shifts the odds. his
outside knowledge adds to the picture of the player. i
have no idea what could be fulfill that function in trading.
it cannot be just the market confirming the result of
one system by a gain and that of another by a loss.
at least i think it can't, because that means you already
banked a loss somewhere. which is unlike the door
switching game. here you do not pay for betting. in
trading you do.

but, MAESTRO, i thought about the same a while back,
when you proposed some trading system, that would
scale up and down pretty much randomly and you
claimed it makes money in real time for you. so i am
humble enough to say that i might not get you ...
 
Think of the Market as a moderator (the host of the show). You have, for example the following combination of the events:

1. A stock might gain $1 a share and then another $1 a share

2. A stock might gain $1 a share and then lose $1 a share

3. A stock might lose $1 a share and then gain $1 a share

4. A stock might lose $1 a share and then another $1 a share.

If you knew that this stock has in fact gained $1 a share (moderator affect) what is the probability that this stock will gain another $1 a share?



Quote from mind:

hmm. the best explanation i ever heard is this. when
you have three doors in the beginning and you pick
one, then the likelihood of having the car is 1/3 and
the likelihood of having not (=it is in one of the two
others) is 2/3. what happens when the moderator
opens one of the two that you did not pick is that the
initial 2/3 is still 2/3, but that probability is now, by the
additional knowledge of the moderator concentrated on
just one door. so you have one with 1/3 chance to win,
and one with 2/3 chance to win.

if you want to claim that a trading problem has relevant
similarity to the 3 door story, there need to be certain
relationships between the systems in place. i struggle
to find the link to trading. it is very clear that the moderator
is the critical element here. his act shifts the odds. his
outside knowledge adds to the picture of the player. i
have no idea what could be fulfill that function in trading.
it cannot be just the market confirming the result of
one system by a gain and that of another by a loss.
at least i think it can't, because that means you already
banked a loss somewhere. which is unlike the door
switching game. here you do not pay for betting. in
trading you do.

but, MAESTRO, i thought about the same a while back,
when you proposed some trading system, that would
scale up and down pretty much randomly and you
claimed it makes money in real time for you. so i am
humble enough to say that i might not get you ...
:D :cool: :cool: :cool:
 
First of all Parrondo and Monty Hall has a major difference.

First, you have a third factor which is the host and the other factor being the choice to switch.

This also leads to the fact that you have a positive Bayesian probability to start off with the Monty Hall Problem. Parrondo's Paradox is a misinterpretation and representation of probability.

They're two different things.

Of course, the only "similarity" is switching... I figured that you used Parrondo as an inspiration and used it differently...

Har har har... I was right...

PS. I'm going to start another gig that won't allow me to post in ET for a while... So... e-mail me... See ya when I see ya.
 
Quote from MAESTRO:

Think of the Market as a moderator (the host of the show). You have, for example the following combination of the events:

1. A stock might gain $1 a share and then another $1 a share

2. A stock might gain $1 a share and then lose $1 a share

3. A stock might lose $1 a share and then gain $1 a share

4. A stock might lose $1 a share and then another $1 a share.

If you knew that this stock has in fact gained $1 a share (moderator affect) what is the probability that this stock will gain another $1 a share?



:D :cool: :cool: :cool:

Thank you for a simple mis-interpretation of Bayesian Analysis. Past results do not count as a "moderator effect". Also, Prediction Group (There was a book written about these guys... I think they're gone now...) did a study on this, google it out if don't believe me me. I've ran similar tests to check how closed and open-position status can serve as a "measure" of choosing what to do.

You're wwwrrrooonnngggg... Again... you're self-interpretting stuff too much. You take an idea and mis-represent the content of the actual material.

Of course, I'm well aware that you've over-simplified your example... but I like to bite... as you know...

Good thing is .... Finally... are we done with Piss-on-doo Paradox...
 
Quote from TSGannGalt:

Thank you for a simple Bayesian example.

Finally... are we done with Piss-on-doo Paradox...

So, what do you think the probability is? :D :cool:
 
Quote from MAESTRO:

So, what do you think the probability is? :D :cool:

Sorry I added more to the content... I thought of letting it pass but I figured I should write more for time remaining...

Over-analyzing your post and relating to my past studies... Would you like me to say... around 60-40 on the positive side...
 
Quote from TSGannGalt:

Sorry I added more to the content... I thought of letting it pass but I figured I should write more for time remaining...

It is extremely refreshing that you are familiar with the Bertrand's box type of paradoxes. But I still insist that you think a bit more about it. Please do not dismiss the premise that easily. It is very rewarding to carefully analyze"well-known" facts again.
Wish you well in your new "gig".
 
Quote from MAESTRO:

Think of the Market as a moderator (the host of the show). You have, for example the following combination of the events:

1. A stock might gain $1 a share and then another $1 a share

2. A stock might gain $1 a share and then lose $1 a share

3. A stock might lose $1 a share and then gain $1 a share

4. A stock might lose $1 a share and then another $1 a share.

If you knew that this stock has in fact gained $1 a share (moderator affect) what is the probability that this stock will gain another $1 a share?

:D :cool: :cool: :cool:
my point is that if you have an effect on a stock that
as forecasting quality, be it from the stock itself or
from exogenous data, you have a tradeable edge. but
out of randomness comes only randomness. systems
without edge cannot be combined to produce edge.
unless (which is a big, big unless) they have certain
features, which i mentioned before.

the monty hall case gets additional data into the equation
in form of the moderator's metaknowledge supplied
for free.

but, well, i did not answer your question. my guess
is 50%. there is no additional information coming
from the fact that it did NOT go DOWN a dollar. the
market is NO moderator.
 
Back
Top