It's YOUR Choice

wyiswyg,

So you think if you lost 5 trades one after another then your chances of winning the next trade increases? Sowell why donot you paper trade the first 3 and trade the next two for $$$ and woho you be the winner...

The above statment of mine is specially true in INTRA _DAY .. This is one of the reasons day trading is so risky.. THE SHORTTER THE TIME FRAME THE NOISIER THE DATA
So we are basically noise traders lol

Just realised I did not address your Second pull back trade.. Please read all my thread.. I have explaind about in efficiencies in the market place.
 
Last edited:
One problem with the balls theorem is that I put them in the bag and know how many there are and what colour they are before I start to draw them out. I also know that there are only two colours - win or lose. What if the bag contained thousands of balls - would they all be one colour or would one be a win - perhaps only one would be a lose?

The chance of being dealt 13 spades in a bridge hand is one in 635,013,559,600. A man dealt such a hand would be staggered, and anybody seeing the hand dealt would immediately expect (probably correctly) some sort of trickery. Yet when we know that the number of bridge hands dealt since the game was invented is doubtless many times the large number quoted, it would be surprising if a hand of 13 spades had never been dealt legitimately.
We could argue until the bag is empty over this one and so will have to disagree. :)
 
Rog,

The number of balls or the colour of them are irrelavant to independency explanation I gave.. If you did not know what was in the bag and the colour of the balls then it would be different scenario but still independcy would be there as long as you did not put the balls back into the bag ..
 
Thanks for the link.. I donot use Fincalc but will later look at their methodology.
My risk analysis is coded into Tradestation along side the VWAP strategy .. As simple as that ..
 
Grey1,

Thanks for the respone. I do have another question. That is, is not the full probability distn of returns EXACTLY the same as running a monte carlo simulation an infinite number of times (or steps).

In other words assume we are looking at trading the DOW in 15 min timeframes. If this is all we are interested in, we already know the past prob distn of returns, so running a monte carlo sim on this will (eventually) only spit out what we already know.

Also, (as another example), the classical demo of a sim in the markets would be to calculate the prob distn of returns over say 30 days using 1 day intervals and then we have many many iterations or possibilities of the way we can get from day0 to day 30. But here's my point again - we already know that the sim is gonna use the 1 day hist prob distn in its calcs.

Have I lost the plot :eek: ?

The General.
 
Grey1 said:
So you think if you lost 5 trades one after another then your chances of winning the next trade increases?
Click to expand...

No, not really what I'm trying to get across. Put it more like this, a stock breaks a cent and then retraces back into its range without going any further. It carries on doing the 100 times, never breaking to the bottom and never more than a point to the top. After that 100th failed break out do you really believe that the next time the odds of the breakout continuing are exactly the same as they were on the first breakout? If they aren't then you can't really see each trade as independant, although as I said I don't think there is necessarily anything to exploit about this it just intrigues the mathmo part of my brain as a concept.

In any case I don't really believe in probability, the outcome of anything is always certain, we just don't necessarily know what it will be :)

wysi
 
General

Quote " Thanks for the respone. I do have another question. That is, is not the full probability distn of returns EXACTLY the same as running a monte carlo simulation an infinite number of times (or steps "

Not quite, as we later be interested in taking the integral of the probabaily function to asses risk and choosing a wrong distribution type will gives us totally misleading results.. I think we have to see what kind of distribution funtion represents cycle in the stocks .. . For example if i wanted to know the probability of MEDI price to lie between the range X1 to X2 with a VWAP of alpha and a SD of Gama then the choice of distibution function would be critical as later I would be wanting to have the integral of that function .. there are many ways of doing that // let me know if you are interested and I will expand on that ...

Quote " In other words assume we are looking at trading the DOW in 15 min timeframes. If this is all we are interested in, we already know the past prob distn of returns, so running a monte carlo sim on this will (eventually) only spit out what we already know. "

Not sure how else you know the probability function of the returns if not using some kind of a simulation .. I appreciate if u expand ..
 
Grey1,

Not sure if I'm digging a deeper hole for myself, but here goes:

Surely if you wanna know the probability that in 1 hr the Dow will be betweeen x and y levels, we can use:

a) the historical 1 hr moves (yielding an average and a certain SD).
b) the historical 30 min move (same as above).

We now have 2 (could use more of course) distribution curves and we can calculate (integral of course (ie the area under the curve)) the probability for each time period that the price will move sufficiently so as to be within the range x and y ?

If the plot has been lost, it is me who has lost it ! :rolleyes:

General.
 
Quote "Surely if you wanna know the probability that in 1 hr the Dow will be betweeen x and y levels, we can use:

a) the historical 1 hr moves (yielding an average and a certain SD).
b) the historical 30 min move (same as above).
We now have 2 (could use more of course) distribution curves and we can calculate (integral of course (ie the area under the curve)) the probability for each time period that the price will move sufficiently so as to be within the range x and y ?


My dear friend , I have been talking about some thing more substantial than above..

By doing above , all we are getting is price is X bar away from Low or High of a pre defined range.. On top of that we would not be able to model the price curve in any kind of mathematical form to take the integral of it .. Also we would not be able to define the confidence interval to calculate the risk of each bar.. You would need Mean price and there is where VWAP comes into the equation ..

Regards
 
Grey1,

Thanks for the prompt reply.

A different angle - what do you have to input into the sim ?
Presumably you need to input the average/min/max attributes for the market moves you are looking at for that particular instrument ?

Is that correct ?

General.
 
Wow folks this is getting very deep. My brain is frazzled!

I get the impression (sorry if this is incorrect), that the consensus is that the probability of a future price can be derived from its past history based on statistics, standard deviation et.al. This is done on a computer so that the trader can focus on timing and entry.

No amount of quantum physics/rocket science or other theory is going to protect you when out of the blue an institution decides to dump block trades on the market in order to depress price. They may do this in order that they can begin an accumulation for their clients at lower prices. Most markets are manipulated by the institutions for their own ends. This even applies to the most liquid markets.

Surely the best opportunities come through experience and understanding - reacting to situations and opportunities when you see them unfold and positioning your size accordingly, rather than letting a computer program dictate what size to take as a computer program cant take price action and volatility in the correct context - previous price manipulation.

I agree though with Grey1's comment about getting in to the trend before others have spotted it inorder to fade those silly/lazy enough to be trading MA crossovers, stochastic ob/os etc. This to me, is what trading is about - understanding the field and what other participants do and how to take advantage of them. While on that point, thats why breakouts 'fail'. They don't really fail at all - they DID break out, its just that the timeframe it was observed in wasn't profitable for the trader. It was really the institutions running the stop orders they know are sitting there! nothing more, nothing less! Another example of manipulation. On an intraday chart, some would call this activity 'noise'!

Anyway, back to this intriguing (and for me educational) MM....
 
BBB,

Quote " I get the impression (sorry if this is incorrect), that the consensus is that the probability of a future price can be derived from its past history based on statistics, standard deviation et.al. This is done on a computer so that the trader can focus on timing and entry.
"

Not quite. No one can predict the future price using any kind of maths, but one can always calculate the risk of taking a trade relative to a price bench mark defined by mass participant .. This is the VWAP price..

Quote " No amount of quantum physics/rocket science or other theory is going to protect you when out of the blue an institution decides to dump block trades on the market in order to depress price. They may do this in order that they can begin an accumulation for their clients at lower prices. Most markets are manipulated by the institutions for their own ends. This even applies to the most liquid markets.

Institutions buy at VWAP or below it they can ., they also donot make up their mind during the day , they execute their trades pre market with a limit order to buy smaller blocks at VWAP or below.. In fact some brokers Gurantee a postion at VWAP for a given fee before the market opens .. I think IB does that .. I stand to be corrected though ...

At the same time a 300K block is not dumped in one single transaction as market maker will rise the price to well above VWAP and execute the sell order in a trailing manner.. if you are a VWAP trader you wont miss the effect of big blocks going through above and near VWAP ..

There is a lot of info in a chart.. one just has to have a logical mind to understand the price behaviour.. ..

I think if you use trade_ideas and set your volume for block trading you are alerted for every single transaction at or above VWAP ..

Regards
 
Grey1-

Your obviously a very knowledgable trader! I must admit, before your this thread, I'd never heard of VWAP, even though I used to be a market maker (Euribor Futures at LIFFE before it went upstairs. In fairness, computers weren't really used in that style of trading. On the floor it's all about seat of your pants and being on the ball, who you know, orderflow and fixing prices with your mates in the quiet times in order to rob paper (paper = brokers)). I'll re-read your posts and see if you reference where I can find more details...

I get the impression that your time frame is a lot shorter than mine. I 'm more of a position trader, sometimes intraday on nothing shorter than a 5 min chart. What you say is very interesting though, especially your understanding of risk.
 
BBB, at first i thought D said 'the nightmare method', ive just used that method on the nasfuts and ive still got an open position and YES! its in negative territory SHIT! ill let you know how i get on, i need it to hit 1409 to break even.
 
You must log in or register to reply here.

Similar threads

H
Forex Scanners
Replies
0
Views
2K
HenryTT
H
W
Home-Based Trading Career Pathway & Journey...
3 4 5
Replies
98
Views
19K
kc8flq
K
barjon
Wot happened next?
Replies
13
Views
3K
charliechan
C
Grey1
Consolidation What Is It ?
Replies
0
Views
303
Grey1
Grey1
B
Trading Timeframes - questions!
Replies
7
Views
3K
Fluke
F
Top