The markets are Random

[Skill Leverage]

This thread amuses me for many reasons

Probably because all of us are wrong, one way or the other. It's a stupid argument to be having (again) - Neo you seem like you have the right intentions, but I think you're massively overthinking this business.

 

[Glenn]

Markets are not random, they are Chaotic (in the maths sense).
Look up Chaos Theory.
Glenn

 

[cr6196]

This is not true at all. Financial mathematics does have many applications in real markets. Banks don't employ entire teams of PhD quants, coding in C++ just for fun. But if you're talking about support, resistance and entries into a market, stochastic calculus probably isn't going to tell you anything useful. And for the record, when it comes to VaR the academics have long told the banks about its problems, but the banks used it anyway.
.

Well it was apparent after 1997, but it is clearly in the academic interest to keep it in banks...i agree though the banks knew the problems but decided not to ditch it as it did allow greater risks and rewards...most of the time. And i agree, the role of quants in entries and exits i dont think is as great as most think...

I just don't believe in randomness cos when you see a trend and people acting the same way (right or wrong)...that just isn't random. And experience and instinct can tell you more than anything else in trading...

 

[Neoripley]

I guess trading is a lot like life itself. There are no guarantees and you never know for sure what's going to happen. The best you can do is stay informed and make sure your ass is covered (life insurance, car insurance, house insurance - all 'hedged bets' /stops in their way) and just go with the flow! Everything in life is a risk and it's normally directly proportionate to reward (or loss). If you don't play you can't lose, but you can't win either. Good luck fellows!

 

[tripleogstar]

exactly

 

[arabianights]

Probably because all of us are wrong, one way or the other.

Pretty much. I mean, you can't even have a 'strategy' for a random market... except for earning the spread, and even that is pretty doubtful. Nontheless, most of the objections to EMH here completely misunderstand the theory (if you can call it a theory; in various statements, particularly the weakest form, it's essentially unfasifiable).

It's a stupid argument to be having (again) - Neo you seem like you have the right intentions, but I think you're massively overthinking this business.

100% agree.

P.S. I still don't think anyone has noticed

 

[Rhody Trader]

Support/resistance or rather market reversals where the market makers/institutions step in, price limits, relative highs/lows, etc.

Don't conflate market makers and institutions. They aren't necessarily the same thing. The objective of the market maker is generally to profit from order flow by selling at the offer and buying at the bid repeatedly. The more the order flow the more the profit. Thus, it's in the best interest of the MM to move its quotes to price levels where the most volume is going to be transacted. That creates a type of market structure. Institutions that are not market makers have other objectives. They are not price creators like the MMs, but they are certainly price influencers in terms of the volume they bring to the market at the prices they determine are important to them. That creates another type of market structure. What we end up with is an overlapping array of structures, all of which sit on top of the basic structure of the market (exchange rules, clearing mechanisms, settlement timeframes, regulations, etc.)

As for your base question about randomness, I'm with Glenn in that it sounds like you are talking Chaos Theory.

 

[Pippppin]

Financial markets regularly exhibit fractal properties - fractals can be both stochastic and deterministic. See the work of Benoit Mandelbrot for an insight into the weird and wonderful world of fractals.

 

[melanie911]

Are the markets structured or random? Is there really any point in market analysis (as we know it) or should someone try to formulate a strategy based on (mathematical) randomness?

I have a strategy based on randomness (fundamental and technical anomalies of the EMH) because I believe the markets are random. But it's a lot of hard work to maintain and to trade it. Works though, most the time.

Do you have a strategy based on market/tech analysis or based on randomness?

ps: By randomness I don't mean totally without structure for e.g. one could just as well assume a water-skier could form a 'moving average' of the boat pulling him. A chicken's tracks in a pen over a centre line drawn in the sand could have 'relative strength' in relation to the direction and proximity to the line. Etc, etc.

you can plan a random entry and random exit, focus'n on position and money mgnt, but you can't achieve real profit when accounting commissions and slippage. I saw similar 3d with test on random trading in other fora. If interested in i can try to retrieve these links. M.

 

[Neoripley]

you can plan a random entry and random exit, focus'n on position and money mgnt, but you can't achieve real profit when accounting commissions and slippage. I saw similar 3d with test on random trading in other fora. If interested in i can try to retrieve these links. M.

That would be interesting if you have it and would like to share that info. p.s. what I mean by randomness is not random entry/exit or random trading. I choose these carefully and follow a tight set of rules but based on mathematical relationship and probablility rather than thinking where should the market go? See, even though I think the market is random I also think there is structure - it's kind of a contradiction, I know, but it's the only thing that works for me (I don't always get it right but I can ALWAYS see where I went wrong). I think it's all based on relativity (perhaps relative strength and all that as well..)

 

[shadowninja]

Random? No. ****ing complex? Yes.

Go back to the origins of trading commodities as in a farmer selling a chicken to another farmer and then expand the market for thousands of buyers and sellers. It's not random at all.

 

[BSD]

Melanie and Neo,

I remembered having read this a while back, scroll all the way down to the summaries at the bottom, you might find this interesting:

Linda Bradford Raschke - LBR Group - Random Entries and Trends

Kinda corroborates what I'm always saying, trading can be just about as KISSy as you want.

Improve on Lindas example above by using timed instead of the random entries she used albeit with a trend filter, ie get the macro direction right like she did, but then wait for the inevitable micro swings against that, and only enter when you're aligning with the macro direction again (macro can be whatever time frame you choose) and provided you stay out of chop-chop and only trade when markets are making strong, smooth moves, you've got yourselves a winning net profitable system.

 

[Neoripley]

Melanie and Neo,

I remembered having read this a while back, scroll all the way down to the summaries at the bottom, you might find this interesting:

Thanks for the link.

Think I'll unsubscribe from this topic now...

 

[Technically Fundamental]

No. You just don't know enough. Do you think everyone in the world who makes money from the market is just lucky?

Maybe the low Tf's are randomish but as long as you know how to read the TF you're trading on nothing is random. I just don't think it's one rule fits all Tech Analysis

 

[Technically Fundamental]

Shocking.

 

[shadowninja]

Perhaps I don't use the word random in it's proper scientific/financial/academic form. Perhaps I should have chosen the word upredictable rather?

In the context of your example I would agree that markets aren't 'random' as long as everyone is buying chickens from the same farmer. But when millions of people of buying chickens, goats, pigs, cows (and that's just the farmyard) from thousands of farmers anything can happen and the aggregate of all that acivity, to me, seems 'random' - or rather, unpredictable.

I stated that you should add loads of buyers and sellers to prevent your comment about "the same farmer" being made - why I bother if people don't read?! It's still not random. It's just a lot more complex. But now you say unpredictable, that is an entirely different matter and needs a new thread!

 

[melanie911]

That would be interesting if you have it and would like to share that info. p.s. what I mean by randomness is not random entry/exit or random trading. I choose these carefully and follow a tight set of rules but based on mathematical relationship and probablility rather than thinking where should the market go? See, even though I think the market is random I also think there is structure - it's kind of a contradiction, I know, but it's the only thing that works for me (I don't always get it right but I can ALWAYS see where I went wrong). I think it's all based on relativity (perhaps relative strength and all that as well..)

the following data are a resumè (no sw to screen shot here sorry):

10 years of daily prices
50 biggest us stocks
random entry random exit
2 to 40 entries per symbol
initial capital $5000
position size $500+commissions

# 30 tests

portfolio (30 test):
best net profit +$3437
worst net profit -$11323
best sharpe +0.28
worst sharpe -1.05

but considering single stock result (30x50 tests):
best net profit +$2412
worst net profit -$2656
best sharpe +1.1
worst sharpe -2.17

so you can trade lucky and random on a single series
and have a sharpe 1.1 that could lead you to suppose your system is robust...

 

[oildaytrader]

Those who espouce the Random Walk type of view of the markets would tell you that the markets are structured but that unpredictable external influences (news items, basically) introduce randomness making the markets essentially unpredictable.

That said, academic research has been coming out suggesting that both low valuation plays and momentum plays can actually beat the market, so market analysis does in fact seem to have value after all. :smart:

These people do not know what they are talking about.

Markets are not always random ,they are predictable when fundamental change.

Here are two examples of currency moves.Are they really random?

http://www.trade2win.com/boards/forex-news-analysis/76816-gbp-fundamental-technical-anylysis.html

http://www.trade2win.com/boards/forex-discussion/76330-audjpy-ready-pop.html

Scotia's reports are showing predictability of markets

http://www.scotiafx.com/Chart_Feed/fxout.pdf

http://www.efmoody.com/investments/efficientmarket.html

O D T