Do you see patterns in Random Walks?

Quote from MAESTRO:

Most regularly!

Well I must admit that you have me baffled...:confused:

Baye's Theorem is usually used to figure out if a "test" is accurate or not. I think some computer programmers use it as well but I don't do that.

I can't figure out what it has in common with trading...?

Any chances you could give me a hint as to what you apply it to...?:D
 
Quote from CoolTraderDude:

Well I must admit that you have me baffled...:confused:

Baye's Theorem is usually used to figure out if a "test" is accurate or not. I think some computer programmers use it as well but I don't do that.

I can't figure out what it has in common with trading...?

Any chances you could give me a hint as to what you apply it to...?:D

Sure:

A very simple (and a very abstract) example. The actual decision making tree is a lot more complex, nevertheless:

Portfolio A has 100 securities in it. All the securities in this portfolio have an average daily range of X%. If any security in this portfolio is at 80% from its daily range (high or low) it has 90% chance of reaching its daily average high (or low). However, if the security 'fails" to reach its daily limit it has 70% chance of "retracing" to its daily open price. On a particular day one of the securities in portfolio A is within 10% of its average high. What is the chance for this security to hit its average high before the day is over.

There are thousands of examples like this and they are all important to consider when creating an AI based decision making algo,
 
Quote from MAESTRO:

Sure:

A very simple (and a very abstract) example. The actual decision making tree is a lot more complex, nevertheless:

Portfolio A has 100 securities in it. All the securities in this portfolio have an average daily range of X%. If any security in this portfolio is at 80% from its daily range (high or low) it has 90% chance of reaching its daily average high (or low). However, if the security 'fails" to reach its daily limit it has 70% chance of "retracing" to its daily open price. On a particular day one of the securities in portfolio A is within 10% of its average high. What is the chance for this security to hit its average high before the day is over.

There are thousands of examples like this and they are all important to consider when creating an AI based decision making algo,

MAESTRO,

Forgive me for jumping in.
When we isolate a company doesn't it mean that the chances will be 50/50 on any given day?

It seems that we also need to know what % of them on average hit the target. A number of days as a sample size would also help.

Although I wouldn't know what formula to use here because I'm not a mathematician. Thank you.
 
Quote from rossky:

MAESTRO,

Forgive me for jumping in.
When we isolate a company doesn't it mean that the chances will be 50/50 on any given day?

It seems that we also need to know what % of them on average hit the target. A number of days as a sample size would also help.

Although I wouldn't know what formula to use here because I'm not a mathematician. Thank you.

Not necessarily. The good example is the Monty Hall paradox or a cancer screening phenomenon. Also, there is a solution for the example provided without any additional data.
 
Quote from rossky:

MAESTRO,

Forgive me for jumping in.
When we isolate a company doesn't it mean that the chances will be 50/50 on any given day?

Of course it does unless you know already the outcome of some other stocks. In the Monty Hall problem, you already know the outcome of the game presenter's choice. This is what changes the probability. But you are correct that at the start of the day, this has no meaning. So any reference to Monty Hall and Bayes is quite unfortunate at this point. Only if trading were that easy...
 
Quote from MAESTRO:

Sure:

A very simple (and a very abstract) example. The actual decision making tree is a lot more complex, nevertheless:

Portfolio A has 100 securities in it. All the securities in this portfolio have an average daily range of X%. If any security in this portfolio is at 80% from its daily range (high or low) it has 90% chance of reaching its daily average high (or low). However, if the security 'fails" to reach its daily limit it has 70% chance of "retracing" to its daily open price. On a particular day one of the securities in portfolio A is within 10% of its average high. What is the chance for this security to hit its average high before the day is over.

There are thousands of examples like this and they are all important to consider when creating an AI based decision making algo,

Reading the question above, I get:
The particular security is at: 80% <= security price < 100%
so:
If any security in this portfolio is at 80% from its daily range (high or low) it has 90% chance of reaching its daily average high
security has 90% chance of reaching high, and 10% chance not reaching high;

if the security 'fails" to reach its daily limit it has 70% chance of "retracing" to its daily open price
Of the 10% that don't reach the high, 70% will retrace to open: 10% * 70% = 7%

So, given the security is within 10% of the high (at 90%), it has the following probabilities:
hitting high: 90%
retracing back to open: 7%
ending somewhere between high and open (exclusive): 3%

Intuition tells me that's not illustrating the concept fully; I think we need the prior probability -- "what is the probability a security will hit its daily average high?" as estimated before the start of the trading day, one step before the conditional probability of 90% given it's at 80% from its daily range.

Unless we can assume "average daily high" is distributed with kurtosis close to 1, then we could assume probability of hitting avg daily high = probability of not hitting avg daily high = 25% (half goes to daily high; half goes to daily low; but that's again with a big assumption of daily price distribution).
 
Quote from MAESTRO:

Sure:

A very simple (and a very abstract) example. The actual decision making tree is a lot more complex, nevertheless:

Portfolio A has 100 securities in it. All the securities in this portfolio have an average daily range of X%. If any security in this portfolio is at 80% from its daily range (high or low) it has 90% chance of reaching its daily average high (or low). However, if the security 'fails" to reach its daily limit it has 70% chance of "retracing" to its daily open price. On a particular day one of the securities in portfolio A is within 10% of its average high. What is the chance for this security to hit its average high before the day is over.

There are thousands of examples like this and they are all important to consider when creating an AI based decision making algo,

MAESTRO,

In this example a piece of statistics was used to arrive at current probabilities. Looking at the point where the security is within 10% of it's high we need another set of statistics to calculate the most probable outcome. But you say the solution here doesn't require new data.

The security now has covered half of the distance towards the potential target and still has 70% chance to retrace to its open price. At this point, I would just cheer them up by adding 5% more so they can reach the finish line that is so close. Although it feels like we are in the grey area and the chance of success is still 90%.
 
Quote from traitor786:

hmm interesting, well i vote that random does not exist. espeically when it consists of data generated by peoples habits, trading style ect..

teh fact that you mention that trend lines do work to some degree shows that what little you know about TA is valid yet overly simplified as most tend to do.

What ever your stratagy is it is based on looking at price movement. whether it be the last 9 bars, past 3 bars or just the bar that has been created. Your choise to buy or sell is based on information to the left of the point where your trade goes through.

Saying that the past has no relevance on the future puts all systems in a drawing boat. Even studies like a moving average are based on the past. they are indeed a pattern. only the easy to calculate ones are being used as they offer i quick simple way to look at a pattern

In a given set of numbers if we find the average and price revolves around the average and may even react to the average this by definition is a pattern. things that happen over and over.

all systems work on this premise.

as mentioned above 99% of these types of traders follow the same old make a channel or connect 2 or 3 dots. and it is inconsistent. Some may find that it bares no coloration.

When we stop to think that maybe we can go further with this and actually entertain the idea , we start to see new things that are relevant.

Surprising to every pattern trader is how past price reflects future price.
Why ?

I dont know , why does price react to moving moving averages?

god knows maybe everyones looking at the same thing.

when price goes in to a periode of consolidation for example, we see it hold that pattern for a certain time frame... why ?

if previous highs can act as resistance and people tend to be predictable ( as our QA class taught us) why does price stay in consolidation for a different amount of time.

news. news is actually incorporated in pattern TA. when bad news comes out it eventually finds some resistance. why at that point ?

is it random ?

or is saying it is random just easier then investigating for hours ?

why is it that when a bar forms 99% of the time it forms where the previous bar was? why dont we see one bar at 20$ and the next bar at 40$ and the next at 5$.

why do people tend to keep price steady. randomness suggest random numbers. Not charts that may look random but all have a bar close to the previous bar.

the charts presented above maybe some guy just frawing a line and assuming that his brain is capable of producing randomness.
but in actuality i see a pattern. every point on the line is very close to the point before it, randomness would have random dots. with no line.

even the numbers in the lottery tend to come out teh same amount of times. some may not come as often as others, but they tend to catch up.

bars stay close together because people have tendencies. they tend to like to buy where price is or around where price is alot of sellers like to sell over price and many like to sell under price.
these are all patterns, orders being filled. buyers being greedy and sellers being greedy and the negotiating. we always buy and sell relative to the last bar we saw. We do not sell 20% lower then where price is on a long.

we let the last bar influence us. as we let the last 2 bars influence us

Show me an example of randomness? it does not exist. you think that you walk a random amount of steps a day ? get one of those things that measures how many steps you take and see a pattern .

here is a chart the lines were drawn before hand as some on here may be able to attest to. I can always post another . these lines were drawn a day before and not moved. we look at connecting points to find resistance. but we dont go further. we would beve imagine that time could also be predicted. how many of the people that were studied for the survey of TA/pattern have lines that intersect and look forward in price and time.


YOu feel free to use the simplest form of connecting pivots to find resistance and assume the rest doesnt work. Why then do you take the simplest form of our science and use it... Cause its easy , and in our science even the easy works a bit.
Do you even know what you are doing when you connect pivots..


you are saying not only that people will react as they did in the past but that they may do so on a slope.

how crazy is that? its like a horizontal resistance line says that people will stop buying or selling at a given price..
a diagonal line indicates that people are so predictable that not only can we judge there buy and sell point but we can judge how they will increase their buy and sell point over a given time frame.


you use that and then you come and give me study saying that its all B.S ?

Look if you dont like it dont use it.

but dont use it improperly like one of your overly simplified stochastics and say its not that good.

people that trade this way have a deep understanding of things. all the studies used are based on patterns and math. you take from us the easy and then complain about its presision.

its like me selling eveytime i get an over bought signal on my study and complaining to you.

do a study on a moving average alone and your results will be the same as a study on connecting highs.

it is talked down even though everything (besides level 2) is based on history. yours are simple and easy, ours take time, we work all night and trade the day. jsut cause you cant do such a thing you cant say it doesnt work.

like any system. if we look on the web it is 99% B.S. actually i find that lines and channels are way more precise indicators even in their simplest form then any other system (besides level 2 )

so go ahead and argue. but we know that you dont have teh computing power to handle suce systems (and no, im not talking about your P.C)

the chart i have included i one of many systems i use, but again it is only one and it was written the day before. no new lines were added after. This is hard to do. look at how many pivot points hit horizontal lines and how many points of resistance hit more verticle lines. we see that we can time in's and outs better with just this one system.. imagine if it was incorporated with your system.

Also i am told that at 2:30 i think it was the news that brought price down... I am man enough to say probably, but what degree did the news effect it was predicted by the lines.


listen i read a book about your system. it was my intention to go down that road. it was not long till i saw that the systems on studies on there own do not correlate with price enough. My 1 system here has a much better coloration and looks out ahead instead of up to the present. sure maybe finding the right studies and combining them may increase coloration, but is that smart ? Is'nt it just trial and error? i there are 50 studies and we need 3 to get the magic combo. the odds of finding it is 50 X 50 X 50

that is 1 in 125,000. and each one would have to be looked at carefully as one that doesn't seem to work when looked at deeper may work.

Why go down that path? as soon as i noticed the chances are slim and i didnt like what i saw right away i stoped reading these damn books. Why did you choose to continue .. did you not realize that the odds were so bad of geting the right combo..

or ...
perhaps
did you decide to connect some pivot points to increase your odds 10 fold of being on the right side of things?

thats fine borrow our useless lines.
but dont come with studies telling me how often you fail at something you dont understand.

do you have trouble getting to the point in a paragraph?
 
Quote from MAESTRO:

Sure:

A very simple (and a very abstract) example. The actual decision making tree is a lot more complex, nevertheless:

Portfolio A has 100 securities in it. All the securities in this portfolio have an average daily range of X%. If any security in this portfolio is at 80% from its daily range (high or low) it has 90% chance of reaching its daily average high (or low). However, if the security 'fails" to reach its daily limit it has 70% chance of "retracing" to its daily open price. On a particular day one of the securities in portfolio A is within 10% of its average high. What is the chance for this security to hit its average high before the day is over.

There are thousands of examples like this and they are all important to consider when creating an AI based decision making algo,

"if any security is at 80% from it daily range." so are you saying that if the distance from the high and low is greater than 80%?

"it has 90% chance of reaching it daily average high (or low)". Are you saying it has 90% chance of reaching one or the other?
 
Quote from rossky:

MAESTRO,

Forgive me for jumping in.
When we isolate a company doesn't it mean that the chances will be 50/50 on any given day?

It seems that we also need to know what % of them on average hit the target. A number of days as a sample size would also help.

Although I wouldn't know what formula to use here because I'm not a mathematician. Thank you.

No, because each company will behave according to maestro's numbers in the long run (this is our "given" info). Sample size won't matter because our domain is all the companies within the portfolio for which we have the relevant statistics for. (remember this is a hypothetical example).

In practice you would be dealing with distributions, so you deal with sample data by having qualified priors.
 
Back
Top