Don't Be Fooled By Randomness

[natureboy1618 10]

...

 

2. TA offers no edge. How can you get an edge from something that cant be defined in exact enough ways to measure?

 

This one keeps coming up, yet none of the TA fans are ever able to provide the evidence of statistical proof that TA offers an edge. There is not one study that has proven this.

 

If TA did offer an edge, trading would be easy due to the ease of learning TA.

 

If all you needed ...

 

TA can offer an edge. Using a pseudorandom number generator to take random entries I recently showed in a talk how using pattern entries can be better than random. How could I do that? TA methods can be defined exactly enough to measure - including chart patterns. I'm a TA guy, so i'm sensitive on this one... i want people to understand its not garbage.

 

Statistical proof? Dude. Read NonRandom Walk Down Wall Street... Statistical as you can get.

 

Trading will seldom be easy.. Even when TA offers an edge. People will screw themselves too often to allow an edge to play out, poor money management, emotionally driven behavior.. Even in automation things can go wrong and poor human decision making comes in to play (Knight).. Many TA approaches are no good by themselves - i'll agree with that.. its like sifting for gold, you can't stare at each unsifted pile and toss it out without even looking!

[BlueHorseshoe 10]

and for the record (again) markets cycle from random to non-random,

 

I agree, they do seem to. But are these cycles you describe random, non-random, or does that depend on some further 'meta-cycle' of randomness/non-randomness?

 

I find the idea of a strategy that discriminates between random and non-random market behaviour uncomfortable, but that's just me. I think it makes better sense to assume that the markets are always either random or non-random (a clearly false assumption), and then to find approaches that balance losses and gains favourably during those periods when this assumption is false. I'll happily explain why if anyone is interested.

 

BlueHorseshoe

[zdo 20]

BHS, ...and when you (happily) explain could you include an explaination in a little bit more depth why you "think it makes better sense to assume that the markets are always either random or non-random" . Thanks. zdo

[DbPhoenix 42]

From a trading standpoint, the practical reality is that it doesn't make any difference. Though I didn't post this series of (6) charts this morning to make any sort of statement about randomness or the absence thereof, it occurred to me when I finished that if I had approached the morning with the idea that it was all random, I likely would have been too paralyzed to take any action.

 

But maybe that's just me.

 

Instead I assumed that traders had some intent, and my course of action was clear throughout the morning.

 

I could have used an indicator, of course, but if I believed the market was random, I probably wouldn't have taken its "signals". To do so anyway would have fed the anxiety that I likely would have felt by not understanding what was going in the first place. This would not have put me in the best place to trade, or not, the remaining opportunities in the day. Or perhaps even the next (trading) day.

[steve46 12]

I think the best evidence (beside common sense) for non random behavior is in today's chart

 

Here you have a chart with three standard deviation lines in place....I base my actions on the concept that a certain percentage of price action is going to be contained within these lines AND I know that other traders make similar distinctions.....the result is that people like myself (people with similar training and backgrounds) make decisions to put money to work based on these concepts....the result is that price is "moved" in a non-random purposeful manner.

 

That's all I have time for today;

 

Best of luck folks

Steve

[BlueHorseshoe 10]

Yes horse i would like to hear you explain that happily.Don't do it grumpily (it's not your style anyway)

 

Do you think it's time to find an alternative to the word "random" ?

 

Hi Mitsubishi,

 

I think of it in the following way . . .

 

Assume that you have a "perfect" methodology for exploiting both random and non-random behaviour. Assume that I have a perfect method for exploiting only random behaviour.

 

Price is behaving randomly, so we're both making money. As price behaviour begins to change to non-random, we both start to lose money. Eventually (unless you can also be assumed to have some sort of perfect zero-lag regime switching model - one assumption too many for me!), you will start to make money once more as you shift to your non-random methodology.

 

You seem to have stolen a gain on me, until . . .

 

The market begins to behave randomly again. From the very instant this happens I am perfectly placed to take advantage of it. You, meanwhile, are waiting for your model to recognise the shift. If the shift is not clean then your model gets whipsawed. My model would never get whipsawed - it's just 'right' or 'wrong'. Also, consider that even if we both make exactly the same, your switching model has expended more 'effort' than mine. So it's less efficient.

 

Even if you have a sound method for identifying such shifts in something like real time, then other problems may arise:

 

1) Consistencyof returns is key to successful money management/position-sizing. In an ideal world, you want all trades to be as similar as possible - the closer every trade is to your avaerage trade, the better (this is why an outlier strategy that has higher single contract average returns as a second strategy may not perform as well as the second when both are optimally position-sized). How likely is it that your returns from each of your methodologies (random and non-random) are truly afine?

 

2) As per my previous post, the rules governing shifts from random to non-random may change, should they even exist in the first place. Most likely, such cycles are also random, at least some of the time . . . Whatever the case, introducing a reliance upon regime recognition into your approach is introducing another potential failure point (I'm sure engineers must have a term for such things). Put simply, the more things you try and respond to, the more chance you have of getting something wrong. If you just position yourself to exploit one type of behaviour, then you have no chance of getting it wrong when that type of behaviour prevails.

 

I hope that wasn't too grumpy, and hope it makes some kind of sense, even to those who disagree!

 

BlueHorseshoe

[BlueHorseshoe 10]

BHS, ...and when you (happily) explain could you include an explaination in a little bit more depth why you "think it makes better sense to assume that the markets are always either random or non-random" . Thanks. zdo

 

Hi ZDO,

 

Please see my rambling reply above . . .

 

BlueHorseshoe

[MightyMouse 12]

 

I think of it in the following way . . .

 

Assume that you have a "perfect" methodology for exploiting both random and non-random behaviour. Assume that I have a perfect method for exploiting only random behaviour.

 

 

A record of randomly generated price and volume points can from time to time appear to be non-random. It would appear to be non-random when it mimics a market generated record of a similarly priced instrument. On the other hand, a market generated record of price and volume, occasionally, but only rarely, appears to be randomly generated. A methodology which trades only random behavior will lose more often than the methodology which exploits non-random behavior if both the former and latter are true.

[BlueHorseshoe 10]

A record of randomly generated price and volume points can from time to time appear to be non-random. It would appear to be non-random when it mimics a market generated record of a similarly priced instrument. On the other hand, a market generated record of price and volume, occasionally, but only rarely, appears to be randomly generated. A methodology which trades only random behavior will lose more often than the methodology which exploits non-random behavior if both the former and latter are true.

 

I wasn't claiming that prices are random (or that they are non-random); I was saying that it may make sense to trade as though they are always one or the other. If your statements above are true, then you could trade as though markets are always non-random, including through those brief periods when they are random.

 

In short, I was arguing for simplicity in a strategy, rather than one which tries to distinguish between two different types of price behaviour (and all the complications that involves).

 

BlueHorseshoe

[onesmith 10]

Price is a derivative of, ....or determined by the analysis of !! .... IT'S PAST CONTENT !!!

 

This constant analysis and revaluation never stops! It is there 24/7 !

 

:-)

[Enigmatics 10]

 

Come on Phoenix. You know everything about everything. Tell me how I made money using a branch of TA where there was no price movement?

 

 

This I got to see!....

 

Ringside seats folks......

 

How to polish a turd coming up.......

 

You talk a big game, but provide a screenshot of this so-called production otherwise I can't take anything you say seriously.

[TheDude 10]

You talk a big game, but provide a screenshot of this so-called production otherwise I can't take anything you say seriously.

 

Bothered.

 

Youve probably read the following posts by now in which case you probably understand a screenshot is really quite irrelevant. Maybe you'd like to do some work yourself rather than expect others to feed you?

 

Anyway - back on topic.....

 

I view the randomness idea as a bit of a fools errand. No one inputs orders 'at random'. Randomness is just what the market may appear to be if you cant make sense or deal with the seeming disorder of the trades.

 

Professional quants treat the markets as random simply as a model as they appreciate they cant predict markets or the next few trades. Just because some treat the market as random, it doesnt mean they think the market is random.

[BlueHorseshoe 10]

No one inputs orders 'at random'.

 

Hello,

 

That's not true.

 

Consider the following time series:

 

2,792,34,4,7,2,567,114,9,1,8,8,441,16,93,4

 

What's the pattern?

 

The pattern is that the number doubles - 2,4,8,16 . . . But you can't see the pattern because I have disrupted it with randomly generated numbers. I know what the numbers are, so I can simply discount them to see the underlying sequence, but you don't know which numbers are random and which are pattern generated.

 

I have added noise to the time series.

 

It has been widely reported that certain market making firms insert orders of random size, at random price, and with random cycle frequency, into the markets to add noise that they can decode but that will disrupt the way that price and volume data is perceived by other particpants.

 

Furthermore, consider that random order placement and random markets are not synonymous.

 

BlueHorseshoe

Edited April 12, 2013 by BlueHorseshoe
It would help if my examples made sense!

[MightyMouse 12]

Hello,

 

That's not true.

 

Consider the following time series:

 

2,792,34,4,7,2,567,114,9,1,8,8,441,16,93,4

 

What's the pattern?

 

The pattern is that the number doubles - 2,4,8,16 . . . But you can't see the pattern because I have disrupted it with randomly generated numbers. I know what the numbers are, so I can simply discount them to see the underlying sequence, but you don't know which numbers are random and which are pattern generated.

 

I have added noise to the time series.

 

It has been widely reported that certain market making firms insert orders of random size, at random price, and with random cycle frequency, into the markets to add noise that they can decode but that will disrupt the way that price and volume data is perceived by other particpants.

 

Furthermore, consider that random order placement and random markets are not synonymous.

 

BlueHorseshoe

 

Are they really doing this or is this just a possibility that someone thought of?

[BlueHorseshoe 10]

Are they really doing this or is this just a possibility that someone thought of?

 

Developers have acknowledged the use of random cycle generators, it seems, but I think this particular application is just informed speculation (it came from an Eric Hunsader article). The idea seems to be to try and screw other participants' read of market depth.

 

Any suggestions as to why else certain firms would spit random orders (and also orders arranged in pretty but pointless geometric patterns) into the market?

 

BlueHorseshoe

[MightyMouse 12]

Developers have acknowledged the use of random cycle generators, it seems, but I think this particular application is just informed speculation (it came from an Eric Hunsader article). The idea seems to be to try and screw other participants' read of market depth.

 

Any suggestions as to why else certain firms would spit random orders (and also orders arranged in pretty but pointless geometric patterns) into the market?

 

BlueHorseshoe

 

I can see random orders being placed and pulled, but it is hard for me to understand why a firm would place random executions unless that firm controlled the bulk of the order flow of that instrument or was part of group of market makers who together controlled the market place. Otherwise, I don't see it and would like to know how or why they would do it.

[TheDude 10]

Hello,

 

That's not true.

 

Consider the following time series:

 

2,792,34,4,7,2,567,114,9,1,8,8,441,16,93,4

 

What's the pattern?

 

The pattern is that the number doubles - 2,4,8,16 . . . But you can't see the pattern because I have disrupted it with randomly generated numbers. I know what the numbers are, so I can simply discount them to see the underlying sequence, but you don't know which numbers are random and which are pattern generated.

 

I have added noise to the time series.

 

It has been widely reported that certain market making firms insert orders of random size, at random price, and with random cycle frequency, into the markets to add noise that they can decode but that will disrupt the way that price and volume data is perceived by other particpants.

 

Furthermore, consider that random order placement and random markets are not synonymous.

 

BlueHorseshoe

 

Sure - you make a valid point. One platform I used allowed the user to specify random volumes in iceberg order types (the child orders are random numbers smaller than the parent order).

 

I see no reason why this wouldnt be extended to time based order types where the times of child orders are selected at random between two points of the day for example.

 

What I meant of course was that the decision to enter an order isn't random. Elements of the execution of large institutional orders may be random, but not the decision itself to trade, or the parameters of the trade. That would be daft of course.

[steve46 12]

These are the kinds of discussions that are (in view) pointless...non-productive

 

For every strategy used there is a relatively simple response that can be used, for example if a participant acting on a daily time is concerned about the effect of HFT on their system they can simply take entries at the VWAP or open an account with a broker who provides that service.

 

Equally simple strategies are available for the intraday participant. It becomes a matter of learning what your options are (educating yourself)...but then success in this business was always about that....

 

to participate on the intraday time frame, I have adjusted the timing of my entries and in the process I actually gained access to additional volatility while maintaining the same risk parameters...and finally, what NEVER changes is the way that professional interests act on the market with respect to inventory bias....they ALWAYS look to squeeze whichever side is out of balance....the result? You can actually use HFT activity to get a nice boost toward your profit targets (assuming you have a reasonably decent system and can figure out where the longer time frame targets are)....

 

What amuses me is that so many here want to debate this at length...without taking the time to really think about what is realistic in terms of usable info....

[BlueHorseshoe 10]

I can see random orders being placed and pulled, but it is hard for me to understand why a firm would place random executions unless that firm controlled the bulk of the order flow of that instrument or was part of group of market makers who together controlled the market place. Otherwise, I don't see it and would like to know how or why they would do it.

 

Hi MightyMouse,

 

Some orders get placed miles away from the last traded price and aren't expected to trade, just intended to spoof the order book, I guess. And some firms probably do control most of the order flow in some thinly traded equities. And some institutions seemingly use iceberg type algos for execution that have randomising elements exactly as TheDude describes.

 

All of this is, of course, completely anecdotal - I have no first hand experience of how a market-maker or large institution operates.

 

Just to progress the discussion, I think it is perfectly plausible to have a profitable strategy that has randomly generated entry orders . . .

 

And I think that people are continuing to confuse the question of whether price is random with the question of whether price is predictable (by which I mean conforms to probabilistic models). The two need not be mutually exclusive.

 

BlueHorseshoe

[BlueHorseshoe 10]

and finally, what NEVER changes is the way that professional interests act on the market with respect to inventory bias....they ALWAYS look to squeeze whichever side is out of balance....the result? You can actually use HFT activity to get a nice boost toward your profit targets (assuming you have a reasonably decent system and can figure out where the longer time frame targets are)....

 

What amuses me is that so many here want to debate this at length...without taking the time to really think about what is realistic in terms of usable info....

 

Hi Steve,

 

The first paragraph above seems useful info - please could you clarify and expand upon it a little?

 

Thanks,

 

BlueHorseshoe

[steve46 12]

Hi Steve,

 

The first paragraph above seems useful info - please could you clarify and expand upon it a little?

 

Thanks,

 

BlueHorseshoe

 

Inventory occurs as a result of transactions. Markets worldwide continue to be primary vehicles for institutional use, with commercial interests adding to that volume...institutions generally "stage" inventory in specific places as they anticipate seasonal price movement (that they generate) over a period of time. Worldwide, we also see Asian and European institutions staging their inventory at specific prices during the overnight market. They do this because they need to generate alpha and the choices available to them are now very limited. The cumulative effect is that an imbalance is created (and acted upon at the next market open).

 

This also relates to the subject of orders placed at distance from current price....again there is little I can say without burning bridges, so I will simply point out the obvious....to the extent that orders (especially size) are placed at a distance from the last transaction, the chances of getting tagged are reduced to near zero, and if there is a utility as a result of that placement (and I can assure you there is) then the purpose is served and risk is managed very effectively using that strategy. In my office orders of this type were termed "smoke" or "smoking"......I leave it to the multi-dimensional thinkers in the crowd to figure it out...

Edited April 13, 2013 by steve46

[BlueHorseshoe 10]

Inventory occurs as a result of transactions. Markets worldwide continue to be primary vehicles for institutional use, with commercial interests adding to that volume...institutions generally "stage" inventory in specific places as they anticipate seasonal price movement (that they generate) over a period of time. Worldwide, we also see Asian and European institutions staging their inventory at specific prices during the overnight market. They do this because they need to generate alpha and the choices available to them are now very limited. The cumulative effect is that an imbalance is created (and acted upon at the next market open).

 

This also relates to the subject of orders placed at distance from current price....again there is little I can say without burning bridges, so I will simply point out the obvious....to the extent that orders (especially size) are placed at a distance from the last transaction, the chances of getting tagged are reduced to near zero, and if there is a utility as a result of that placement (and I can assure you there is) then the purpose is served and risk is managed very effectively using that strategy. In my office orders of this type were termed "smoke" or "smoking"......I leave it to the multi-dimensional thinkers in the crowd to figure it out...

 

Thanks - this type of post provides a perspective I don't have.

 

Just to be clear, are you saying that short term participants will "squeeze" longer term participants, or visa-versa? Is there any kind of integrity/consistency to the motives of long term participants upon which such strategies could be realised?

 

BlueHorseshoe

[BlueHorseshoe 10]

These are the kinds of discussions that are (in view) pointless...

 

What seems pointless, I debate at length; what seems point-ful . . . I just get on with.

 

That way you never risk throwing out a gem with the trash ;

 

BlueHorseshoe

[TheDude 10]

theres only one reason why orders are placed away from the market ahead of time - and thats to get time priority in the queue. if you dont need/want to trade at the level, then you pull the order.

[BlueHorseshoe 10]

if you dont need/want to trade at the level, then you pull the order.

 

Within reason/ability to pay fines, yes.

 

We're re-treading old ground here though, and it doesn't have much to do with whether or not price movement is random . . .

 

Both short term and long term participants react to price. I would contend that price movement is more random in the shorter term. Which must come first, the random price movement to which shorter term participants react, or the misguided reaction that leads to random price movement in the short term?

 

And are longer term price movements really as non-random as they seem? If your outlook is "forever", the stock market always seems to trend up. But what if you enlarge your perspective proportionately? What will an equity chart for the past 100 years and the next 900 look like?

 

"Prices always revert to their mean, and over a long enough timeframe that mean is always zero"

 

BlueHorseshoe

 

BlueHorseshoe