i want to understand this

G

Gordon Gekko

forget real trading, i just want to understand this concept.

let's take a truly random event, meaning the outcome of the last trial has no effect on the current trial.

say there is a stock that is either up or down and the probability is always exactly 50/50. is there any difference between going long when there are 20 up days in a row (uptrend), 20 down days in a row (downtrend), or 20 days of alternating up and down (chop)?

i would guess the answer is no.

i know i've mentioned this post a few times, but i think it's a great question:

Originally posted by jperl
...here are four NASDAQ stocks whose number of updays (close to close) over the past 100 days is very low:

percent updays
LRCX 37%
TMPW 37
AAPL 36
LRCX 34

Question: Would this data influence your decision to short these stocks at the close of today? Or if your system tester suggested a long entry tomorrow, would the above info influence your decision not to take the long?
More...
Originally posted by Gordon Gekko
Traden4Alpha is right, this has been an overlooked post, imo. is the answer to this question yes or no? if no, that the data should not influence you, why look for trends then? i think most people here try to identify a trend in some way, with moving averages or whatever. could you not say the 2 are the same? if that data of % up days does not matter, why would price related to a moving average, or the slope of a moving average matter either? or, is it true, that the above data SHOULD influence your decision?
More...
as i said in my above quote, if you should not make a decision based on information like that, what's the point of using the slope of a moving average? or what's the point of looking for trending price?

some people have said to have 2 systems, one for trending and one for nontrending. but what is the point of that if you have no idea when one is going to start or end? if looking at historical prices doesn't matter, why would you switch to a chop trading system when you see chop? it could just start trending after you switched. so then you switch to a trending method and chop could occur! i want to know if looking at previous price movement should influence your decision at all for how you would go about making a trade.
 
The only difference between up-days and down-days that I can see is that after a series of down-days the relative size of a 1-point move would be greater than before, i.e. stocks can't fall below zero. If you lift that restriction, I don't see any difference.
 
Originally posted by Gordon Gekko
forget real trading, i just want to understand this concept.

let's take a truly random event, meaning the outcome of the last trial has no effect on the current trial.

say there is a stock that is either up or down and the probability is always exactly 50/50. is there any difference between going long when there are 20 up days in a row (uptrend), 20 down days in a row (downtrend), or 20 days of alternating up and down (chop)?

i would guess the answer is no.
More...

Hey Gordon...I think the key word in your question is "random".

If you believe the markets are random, then you're right,
the answer is no.

But, if you believe (like me) that the markets are not random,
then your chances are not always 50/50.
 
Originally posted by Gordon Gekko
forget real trading, i just want to understand this concept.

let's take a truly random event, meaning the outcome of the last trial has no effect on the current trial.

say there is a stock that is either up or down and the probability is always exactly 50/50. is there any difference between going long when there are 20 up days in a row (uptrend), 20 down days in a row (downtrend), or 20 days of alternating up and down (chop)?

i would guess the answer is no.
More...

The answer is no. As an example you can look at the roulette tables in Las Vegas. They did not build those huge Casinos by allowing people to figure out ways to beat the system. Basically you can hit black or red and then there are two green numbers. By removing the greens the odds are exactly 50/50. Black could come up 50 times in a row and on the 51st spin of the wheel the odds that black come up are still 50%.

What most people get confused on is they see that black has come up 9 times in a row and figure the wheel has to stop on red pretty soon. The odds of black coming up 10 times in a row is almost 1 in 1000. People look at this and actually use this to determine their bet. In reality it is a sucker bet as while the odds are 1 in 1000 that black will come up 10 times in a row, the odds it will hit black on the next spin are still 50%.
 
Originally posted by feens93129
What most people get confused on is they see that black has come up 9 times in a row and figure the wheel has to stop on red pretty soon. The odds of black coming up 10 times in a row is almost 1 in 1000. People look at this and actually use this to determine their bet. In reality it is a sucker bet as while the odds are 1 in 1000 that black will come up 10 times in a row, the odds it will hit black on the next spin are still 50%.
More...
that was a good explanation...thanks.

will you take the next step and try to explain the stock market? ;)

by that, i mean...if you were to find the #1 stock in the nasdaq with the most number of down days of the last 100, would that influence you to go short? would it make you hesitate to go long? what i want to get at here is... in the 50/50 example, we know what the probability is. how do you handle the stock market? if we figure out what % of the last 100 days are down, can we get from that some type of probability for future trades? or are past results not indicative of future results? would there be some value or no value at all? if there is no value at all to that, what kinds of entries can their possibly be that are actually better than random?

this will lead me to my next question... if we can say that previous price movement does not necessarily predict future movement, why do people like to find trends? could they not just end as soon as you get in? same goes for chop....say you have 2 trading systems, one for trending and one for chop. what is the point of switching systems when you have no idea if chop or trend is going to happen next?

for the record, i do not believe the markets are random. as i've said previously, i think markets are chaotic systems. that being said, how the hell do you deal with it? i've learned to not make predictions. i also accept that anything can happen at any time. however, after knowing this, what can you do? what kind of entry can be good if we have no idea what is going to happen? i know you would go with the highest probable entry, BUT HOW DO YOU DETERMINE WHAT IS HIGHLY PROBABLE WHEN PREVIOUS PRICE DOESN'T PREDICT FUTURE PRICE? as in the example of the stock with many down days, if we can't use that to find a good short entry, what CAN we use to find a good entry?
 
GG - Re: Trending

I think the answer is this. No one can really say for sure that the market "is in a downtrend" or "is in an uptrend".

All someone can say for sure is that the market "was in a downtrend" or "was in an uptrend" the last time they looked, which, of course, may have been 30 seconds ago.
 
Originally posted by Quah
GG - Re: Trending

I think the answer is this. No one can really say for sure that the market "is in a downtrend" or "is in an uptrend".

All someone can say for sure is that the market "was in a downtrend" or "was in an uptrend" the last time they looked, which, of course, may have been 30 seconds ago.
More...
quah, yes, i agree totally. but knowing this, what can we do? i'm glad you replied, because i can use your trading system to make my point. btw quah, i'm glad you're here because you seem like you've thought about the same things i do...and you kinda understand where i'm coming from..i think.

anyway, quah, when it comes time for you to enter a trade, you are basically going with the trend by looking at the stochastics. at the moment you place your trade, you are basing it off PREVIOUS price movement. you're basically saying, because price "WAS in a downtrend" when i looked 1 second ago, i'm going to go short.

what is the difference between that and finding the #1 stock in the nasdaq with the most down days of the last 100 and shorting it? some would say this isn't a reliable way to find an entry.

btw, quah, i'm not questioning your strategy. i think you have a great idea. i just want to get to the nuts and bolts of how profitable traders are actually making money. to me, the answer is not "make 3x your losses" or "have a positive expectancy and follow your plan." that may be technically what is happening for profitable traders, BUT HOW ARE THEY ACHIEVING THOSE RESULTS? if i want to be a profitable trader, i can't just magically get winners 3x my losses. i want to know HOW they are actually doing it.

before people tell me i have to find out on my own, look at rtharp. he will flat out admit he tries to mimic other profitable traders. he will say he doesn't want to reinvent the wheel. i think it's wise to be like that. i've already lost a lot of money and i need to change how i'm trading, obviously. i want to find out WHAT people that make money are actually doing different from me. the philosophy of trading is TOTALLY WORTHLESS unless you know how to implement the principles to make money. i'm proof! i know all the principles but i'm not implementing them in a way that makes me money. i want to know what profitable traders do differently than 90% of everyone else.
 
from Gordon Gekko:

say there is a stock that is either up or down and the probability is always exactly 50/50. is there any difference between going long when there are 20 up days in a row (uptrend), 20 down days in a row (downtrend), or 20 days of alternating up and down (chop)?

i would guess the answer is no.
More...


I see that someone wasn't paying attention in school. :)

The answer to your question above is it depends on the degree of autocorrelation of the tradable.

Take Bonds or Currencies for example. They tend to exhibit to high degree of non-randomness (high autocorrelation). Therefore in that case, yes there is a statistical edge in going long on the 21st up day in a row.

Stocks on the other hand, have a much higher noise-to-signal ratio. A time series of share prices tend to exhibit a high degree of randomness. And as follows there is no edge in a certain position based on previous data points or on the close of the previous day's range.


Make sense?


Dr. Zhivodka
 
Zhivodka, this is probably a different thing I'll mention but I read somewhere that different markets have a different # of "chaotic dimensions" or whatever its called i forgot. currencies have like 3, stocks 7, something like that. the more dimensions, the more complicated the market. do u know what im talking about? i thought you might.
 
You must log in or register to reply here.
Back
Top