why have you accepted the price-predictability assumption?

Hi Turveyd, sorry but that was not the point either.

Aha. I only said predicted as in the thread title. It is a lot easier to listen to what someone said in the past and then make the pieces fit in the future. Back to thread. It would be good to see outcome of generated charts for e/u against g/u and e/ch. I personally do not thinm they could be viable

Yeah it failed, good job I wasn't risking any hard earned pennys, it was never worth risking anything over the festive season.

I appreciate the post. You'd be amazed at how well you can view the evolving structure. News, fundamentals, TA, often speak a very different picture. Its somewhat complex, yet its a very simple concept. I had to literally forget everything I had ever learned about TA. Once you grasp the concept, you'd be amazed how simple it is. Once you do analysis, all you have to do it monitor the market for change.......if you're interested and need any help looking for any info or resources you know where to find me.................

Very interesting thread.unusual practice here on ff.

There are some very good posts here.and I would say if you could Tie them together you would have a rewarding trading methodology.

The question of random verse non random can only be truly answered if you have found a way to calculate these variables in a different way.i 100 % believe the market to be non random.the unknown variable as to futer events being the only factor that remains random.but that applies to every last thing on this planet.we factor this unknown variable into our lifes to minimise its affect.no differently to money managers who employ hedging and bonds futers ect.

You can clearly see the markets non random behavour through the ATR and Volatility .but it not so clear from one tf and one pair.the problem being the calcs are flayed.also the volatility and atr need to be calculated backwards.from lower tf to higher.the 3rd variable required for the volatility is cross market calc.1 pair will give randon results it neefs to be calculated with correlations to the related currencies this will clearly show non random movement from random range.

The market smoothes out if calculated using more market data.

If someone was able to show volatility and atr correlated and calculated from a fixed tf but using the data from lower you would see pure structure.

I see the spread mentioned and dismissed.both points are relevant.the pipletts are pointless.but the way the spread opens shows the immidiate bias.but thats a very different topic

I also stand corrected.

I defended my point regarding any successful trading method required one element of discretion to be successful longterm.i was wrong.using the calculation s mentioned in my previous post it is possible.

back to my Lsd.

So , Is it going up or going down?

I assume you have done your stats homework since you state it as a conclusion? If yes can you expand on this please? Thanks.

Ok nobody here said its efficient entirely.

I myself belive to be either:

1. pseudorandom, so the structures repeat themselves, but between them are random, just by definition, so that patterns could rise, and the fact that we can trade patterns means only that the "seed" of the pseudorandom algoritm by which the market is built up changes just at that moment when a pattern rises, and also the seed changes very frequently, i belive weekly, so 1 week from monday to friday has 1 seed, and the next week will have another one, because in the next week the traders and other market participants will have a different attitude towards the market so its logical that the seed changes
2. or it is true random, but the distribution of up /down moves are not 50-50, thus sometimes it gets influenced by external factors (economic factors & financial factors, interest rates, interbank rates,etc...)

Now i could not decide which one it is, but i hope i can find the answer here in this thread.So keep up the good work guys !

But if someone assumes its true random with 50-50 probability of up/down moves and that it doesnt have a bias a.k.a efficient market hypothesis, then they are incorrect, because:

1. Repeatable patterns exist, no doubt, which repeat themselves too much to be just a coincidence
2. They are profitable if you know to forecast them (or atleast give the probability of it)
3. Structures are forming on the larger scale
4. Many scientists agree about auto-correlations in the price
5. Some patterns may appear randomly, which are not influenced by any fundamental or other event, but there are still some patterns which appear at a regular basis (my "U" shaped pattern discussed a few pages back,pennants, flags,head & shoulder, double top,etc)
6. If it were to be true random with absolute no bias then nobody could make a dime on in,therefore nobody would trade the market,therefore the market would not exist, a paradox here

---------------------

Now responding to your post mister me , actually a volatility meter is very wise idea, but i dont think it will always help.Just looking a few pages back i posted a nice and innovative trading method, but it got dumped by some troll posts afterwards, so let me requote myself:

http://www.forexfactory.com/showthre...76#post7179476

Especially check the picture, if the price were to be in range just like there, and a breakout were to be pending, it has a higher probability that the price will break out in the direction where it was closer to 1 range than the other.You can clearly see that in the picture the price was right on the upper range, so just if like the price was screaming : HEY FOLKS I WILL BREAK THE RANGE UPWARDS!

So if we know that if the price is closer to 1 range than the other the probability of breakout in that direction is higher than in the other, if we were somehow to guess when will that breakout happen, we could make a lot of money there.

So its important to check the volume buildup, higher volume = higher ticks/second = more traders enter, which is autofulfilling prophecy, since nobody knows when will the breakout happen so everyone jumps in the trade, but the mere fact that everyone jumps to trade will cause the breakout

So in my point of view the volatility measurement is crucial to breakout strategies, now i dont know if the ATR is the best tool for that, but the volume is.So we would need a volume + price distance over time type indicator of some sort if we want effective breakout trading.

Honestly this is my view, what do you think about it mister me ?

You don't have to mathematically prove the market is random, just observe that it is unpredictable.
Two entirely different things.
You can predict the market is going up, and keep predicting that it will go up until finally it does. Your prediction rate is 100% successful - but is it of use?


within any unlimited random series of digits (0 -9) there exist longer and shorter series of digits which don't appear random at all. Well,are they?

I dont know if actually you were adressing me with that post but i will respond if you dont mind.

Well we dont know if its random or not, we can never know its true nature, but we can observe only what we can sense, and the mere looking only suggests that it is pretty close to random, even if its not its hard to tell.

Its just like the fact that how do you know if a wall is red colored if the room in which the wall is its completely dark ? The same way, you cant know the true nature of the market, but rather only know the part which can be detected and observed.

And we have to separate 2 types of randomness: pseudo and true, as i was talking about in my last post.It can be only 1 of them:

1. true randomness doesnt allow patterns to form, but if its a biased, meaning that the distribution of candles is biased in 1 direction, so that for example a trend, more upticks than downticks, could give rise to patterns even in a true random enviroment.
2. pseudo random in itself can be a pattern forming because some stuff can repeat itself, some structures which are so familiar = chart patterns, so even if the underlying market is pseudo-random, patterns can repeat (this can be also biased but not necessarly)

So its hard to distinguish which one is it, my bet would be pseudo, because the sentiment always changes,which could act like a new seed for the psudo algoritm by which the market is made the matching orders system.Of course the sentiment could change on the big scale and on the small one, creating an even bigger confusion, but still a pseudo is a pseudo,thus this allows patterns to form.

We also need to examine what moves the markets, are they the speculators, probably but not limited to: mostly big banks & hedge funds
The bigger part of the picture is the international economy,which is a bigger "motor" of the market and the fuel is the entire money supply of the Earth since all markets on Earth are connected in one way or the other.

Secondly a prediction also have to have a time interval, you didnt think i was so stupid to predict something and not associate a time interval in which the prediction should occur

I could say the E/U will go up, and i could have a 100% win rate because its certain that it will go up at some point, the question is when? Without "when" associated with the prediction, the prediction is pretty much invalid because i still have to enter the trade at the optimal moment.

So its wise to segment the market into pieces, if you are an intraday trader then probably 1 day is your interval, so you just have to find a way to predict, is the E/U going up today or down? If weekly trader then on the entire week and so on.But thats not the point here.

Because my theory is not like this, if i assume pseudo randomness of the market,then the market is still random in a sense, so its still impossible to predict the price just like that, but the point is not to hit the market where its strong but where its weak.

So i will not predict what i cant, but i will predict what i can, and since pseudo randomness will inevitably repeat itself,thus forming patterns,hints of good trades and market ineficiency,chart patterns, i will concentrate all my trading abilities to find the patterns which do repeat in the vast randomness of the market, AND not waste time to predict the entire picture which is pretty much impossible.

Basic Fundamental of AMT price + time = value
Because the market has been shown to not be efficient; so, theoretically at least, timing, denied by efficient market theory, can possibly be profitable.
Time validates price (price over time = value), and Time regulates opportunity.

That’s why the profile indicator is used for viewing “price over time” of various TF’s
It’s called the Auction Process. This is a 2-way Auction market in which we trade……..
You can’t expect to “Beat” an auction market unless you know “what” it is & “how” it operates..........

I could say the E/U will go up, and i could have a 100% win rate because its certain that it will go up at some point, the question is when? Without "when" associated with the prediction, the prediction is pretty much invalid because i still have to enter the trade at the optimal moment.


This is why I shot down Magix after he had a pop at me, he doesn't use SL's, so there are no Time restraints on his trades, most will work, but ultimately 1 will also wipe him out 1 day sooner or later.

It's easy to attach Random, to something you don't understand, which is the main issue, I understand it and it's quite simple when you do understand it, and it's far from random.

Being not random, doesn't make it any easier to predict, just because you can't always predict it, doesn't mean it's random, but because you can at times predict it means it's definately not random.

Simples!!

And I think that will do me on this subject, I'm outta here, enjoy!!

Since you are gone, you will not see this.....but, to those who are following, you can use a no SL strategy, if you are an active follower of the market. The answer is simple. I have said it here several times. When price is going up, buy. When price is going down, sell. At some point, you will have an imbalance of buys and sells. If you are patient enough to wait for this, you will win, each and every time.

And you have to drag me back in!!

You forgot :-

" IF you trade small enough and don't get to a margin call before you get to profit "

Predicting volatility might be useful for trend-following systems, which I'm working on right now. And you don't need to fit so many lags, from my experiments, 100 lags is enough, from there on you don't get much better predictability.

Well said Proximus:

The market does indeed have characteristics. When price is going up, it behaves in certain predictable patterns. When it goes down, likewise. When it goes sideways, This is also true. The problem is that when it changes, it does so at random, more or less, and more or less without warning, appearing to be in one type of behavior while actually in another.

It is an artifact of human behavior, and they always have shapes and patterns.

When an archaeologist first digs into a city he may not always be able to tell you where it is, but he knows of a certainty there is a junkyard in there somewhere...

What I have not observed ever is a price gap of over 50 pips on a one minute chart while the market is in session. If humans do not do anything to it, the market price remains right where it is.

Now human beings watch one another. Always have, always will. When some decide to buy, others are tempted to follow when a lot of people are buying,and the price is moving up parabolically, sellers tend to be inhibited from entering the market. Until a point is reached when enough sellers are convinced the price is too high. Then it swings in the opposite direction.

Price tends to be a continuous function of time, and tends to mimic the growth pattern or feeding pattern of living things.

That is why Fibonacci sequences appear sometimes to be effective ( although somewhat too simplistic, better than nothing) , because the medieval Italian mathematician contemperary of Da Vinci was trying to build a mathematical model of the growth of living populations.

One thing I try to do is draw lines of resistance and support from areas where the market hesitated before. I try not getting in the market at these points but waiting to see if it blows through them. or hesitates, or even reverses there. I wait for the market to make up its own mind, try to ascertain, which kind it has decided to be, and get back in accordingly.

None of this is certain. There is only less uncertainty in some directions than others.

Good luck with that Grail you are building, though to me it seems to have more and more of the shape of an ark...

May I ask what is your definition of "volatility"is?

Volatility is simple in concept, surprisingly complex in it's determination. Gordon Gustafson in the June 2001 Stocks and Commodities defined the short timeframe volatility as the ATR (a days true range). Kevin Lund in the May 2002 Stocks and Commodities defined volatility as "rate of change in price over a specific time period". Both err. Volatility is an (average) price fluctuation, not a velocity. It is created by differences in opinions about value. Mixed in with a volatility measurement is random noise, a fluctuation due in part to the tick size. If a market is in a trend, volatility (and noise) rides along, helping to obscure the change in value.

Option traders often use implied volatility, a value that is backed out of the traded option price. Since price itself contains noise and other fluctuations, implied volatility includes "real" volatility plus an unknown fluctuation factor. Implied volatility is prone to all those errors associated with price variation (noise, manipulation, large trader's strategies, etc). Long Term Capital Management which lost $1.3 trillion, had its largest positions in implied volatility trades (R. Lowenstein, When Genius Failed....., Random House, 2000).

A trader needs to know yesterday's volatility and the day before, not the average for the year. Changes in local, day-to-day, volatility alerts the daytrader to potential changes in price.

I actually tried this for about a year when I first started . Went from a $4000 account to $150 before I abandoned it. Back up now to over $3000.00
Not interested in trying it again. When the debt you carry is dragging your account down with it, it is time to consider closing at a loss. Some would say , well past time.

No single strategy , however reasonable seems to work in all types of Market "Weather". You have to learn to trim the sails.

You didn't try it on a low enough risk rating then I guess, ofcourse if your not already rich then the returns really don't make it worth it with that low a risk, maybe if you want 10% per year, but not nothing to wage.

Choppy ranged markets it's fine, get caught with Longs on GU on that move from 2.1000 to 1.3500 area and well, doomed.

Looking like my new method, is simple enough to work in all market conditions but we shall see, if only someone would of said to me put 2 envelopes on a M1 chart with these settings 7years ago hey LOL

No interest in the Random or not debate, my mind is made up ( NOT 99.882% sure ), so no point looping around for ever on that.