Are currency markets random?

[TraderABC]

Hi all.

Are currency markets random or do they have a non-random element. If so how much? TA signals are rarely super profitable, and most of them are too similiar to each other. The fact that some traders make money is not a proof of market predictability. There are millions of traders, some of them may simply get lucky.
Clicking buttons is not laying bricks in -40 weather... Very often this or that "Guru" tells us about this or that secret method from Japan centuries ago or whatever... Well anyone can put their signal on a chart and find a chart that shows how the method works well and shows that the price is not random. It is unprofitable for people to sell their systems if market is random, plus it hits your pride knowing that your success was due to luck... Even if the market is predictable, ordinary people like us probably cannot workout the formula for prediction and those who know will not release it. If there is a 100% can't fail model of price action if enough people use it it fails....

Any way to statistically measure if the market is random or not? I've heard something about hurst measurements. Am I supposed to use O H L C or OHLC/4 ? How many candles should be there to form a statistical answer?

Thank you all. I hope that market is not random and people can make money there...

[socrates]

According to respected authorities, the markets are non-linear, dynamic systems. Chaos Theory is the mathematics of analyzing such non-linear, dynamic systems. Chaos analysis has determined that market prices are highly random with a trend component. The amount of the trend component varies from market to market and from time frame to time frame

[pqt]

I'm no expert but from what I can see, short time frame movements tend to be much more random than long term ones, probably because of influence of rumors, technical blips, big players interventions, etc...

[tommyfx]

HI trader ABC,

I have gone through times when i felt the market was random and lost all hope that trading was a skill you could build up. But i have learnt that the markets ARE beatable i consistently have profitable trades and so do many people, you don't see major companies hedging their yearly import costs on the roulette wheel, and the reason being the roulette wheel IS totally random, 50 reds in a row doesn't mean the next number is any likely to be a black etc etc But financial markets do have certain factors that mean they are predictable. MArket price is a result of supply and demand, obviously the more people demand a currency the higher the price will go, to move the forex market requires HUGE amounts of money, million pound positions are nothing to this market, so for the the EUR to make the gains it did against the dollar in the year or so leading up to about April required a lot of people to make a judgement on the value of EUR/USD and simultaneously agree that the EURO was undervalued, this isn't an action of a random market.

I would add, however, the smaller the time frame you are trading off the the more "random" elements of supply and demand creep in. The shorter time frame traders are in a constant battle over supply and demand, when you have traders looking for 10-20 pips per trde, and professional institutions market making looking to take 1-2 pips at a time, it is unlikely you will have any sustained market momentum to create a predictable trend.

I suggest bringing up an hourly chart or even a daily chart and you will see how the market is nearly always in a state of a majority agreeing the market is over valued or undervalued and my advice is agree with them.

Hope this gives you a bit of faith, and if i am wrong and the markets are just random movements resulting in people randomly entering trades then who am i to argue? I am making lots of money so must be doing something right.

Tom

[slope30]

Tommy,

Just a couple of questions. What system have been working all this time and have you heard of J-charts? The J-chart system seems to use chaos theory and fractals. I am looking for someone who has extensive use with them.
-Thanx B

[Gamma_Jammer]

I would say that the forex markets are definitely, unquestionably NOT random, even in the short term. This largely stems from the fact that they are not a pure speculative market, and as such not a zero sum game in the truest sense of the word.

Things like daily fixings etc are demonstrably not a random phenomenon. However there's a huge difference between saying something isn't random, and saying it's entirely predictable (which I am certainly not claiming with respect to the currency markets). But I definitely see patterns occurring every day that are non-random.

A few examples;

Central bank intervention in the markets at certain levels;

Daily fixing trades (particularly the midday (London time) Frankfurt fix and the 1600 (again Ldn time)WMR fix)

Daily and weekly squaring up activity (watch how often after a big move on a friday (e.g. payrolls related move), the U.S. evening session will show a steady move the other way, as bank spot desks, futures punters and other short term names square up ahead of the weekend. You also often get movement when the US stock markets open and close (less pronounced than it was a few years ago though).

I could probably list more, but hope that these few have demonstrated my point.

GJ

[tommyfx]

slope30,

I have heard of J-charts although i don't have "extensive use of them", in fact my only contact i've had with them is general research into trading.
They do work partly on the theory that the markets have random elements (when the herd mentality forces people to aggressively buy/sell a price and greed and fear take away the hunt for fair value making price movement impossible to predict) this is my interpretation of the following extract from a trading encyclopedia on J-charts,

"Price Forecasting
J-Chart treats markets as energetic systems, thereby giving us a new way of looking at them. It is designed to help the trader decide when markets are in equilibrium and when they are not. The closer the price action comes to filling a perfect isosceles triangle in a given period (turned on its end), the more it is in equilibrium.

If markets were efficient, they would also be logical. But as any trader knows, markets are neither totally efficient nor totally logical. The reason is simple: markets are prone to the herd mentality. Herds rarely move efficiently, and they are certainly not driven by rational logic. They are more likely to vacillate between periods of greed (when prices are driven up in the rush as people buy not wanting to miss out) and periods of fear (when people realize they got carried away)."

Looking for equilibrium after periods of greed or fear? (an aspect of supply and demand, which wouldn't be present in a random market, surely?) Seems to me this is a way of finding levels when the market is overbought or oversold, which again is a concept a random market wouldn't understand.

As for your second question, what system am i using, it is a trend following system, on an hourly chart, its really very simple, based around moving averages, pull up an hourly chart and find an indicator/s settings that allow you to enter a trade "near" the begining and exit "near" the end of a trend and you will do just fine, too many people become obsessed with predicting the market and catching every pip a trend has to offer. From my experience getting 50% of a trend 90% of the time is far better that getting 100% of the trend 10% of the time.

Good luck,

Tom

[Medran]

Originally Posted by TraderABC

Hi all.

Are currency markets random or do they have a non-random element. If so how much? TA signals are rarely super profitable, and most of them are too similiar to each other. The fact that some traders make money is not a proof of market predictability. There are millions of traders, some of them may simply get lucky.

...If there is a 100% can't fail model of price action if enough people use it it fails....

Any way to statistically measure if the market is random or not? I've heard something about hurst measurements. Am I supposed to use O H L C or OHLC/4 ?

Hi Tom

Price action is made up of signal (the market telling you what it is likely to do next) and noise (the random bit). The Signal/Noise ratio gets greater as you move down the time frames and in (for example) a 30 second bar there is virtually all noise and no signal whereas on hourly and daily bars the signal/noise is good enough to trade (some might argue 15 min or even 5 min bars but I would differ).

> TA signals are rarely super profitable, and most of them are too similiar to each other

They are just tools. It is how you use them that makes them profitable. You'll find the breakout type signals work best when noise is high (on short time frames) and that trend type signals work best on longer frames. You can group just about any indicator into one or other of these two categories.

> ...If there is a 100% can't fail model of price action if enough people use it it fails....

I believe this. That's why you need to develop your own system rather than get one out of a book.

> Am I supposed to use O H L C or OHLC/4

Use what you feel is appropriate. I use C, H+C/2, H+L+C/2 and H+L+O+C/4 depending upon what I am trying to do and the logic behind the trading rules I am testing. A straight C works best in very short time frames as it is closes t to the action and a H+L+O+C/4 might be good for picking up smooth trends in longer averages and longer time frames.

It is possible to make good consistent profits out of Forex and this bunny is worth chasing (though you may have a long run ahead).


Have fun


Medran