A Theory of Market Action: Part IV - The Theory

[madspeculator 10]

@pmwhite, @gooni:

 

I concur with BlowFish's answer to your questions.

 

@gooni:

 

Further, BlowFish's comments and my response to his comment might provide you some clues as to why all market order buys at the ask are not necessarily "buys".

A small hint: By assigning all market order buys at ask as buys, one will be double counting buys! Enough said. As I said in my reply to BlowFish, BlowFish might be on to something!

[madspeculator 10]

Is it correct to say that CVD is measuring the liquidity takers (commercials?) as the market order takes liquidity out of the Limit order book pool. But once they took the resting limit orders, they are left with the inventory which happens to be the liquidity provider once the market is visiting this zone afterward

Karish

 

Karish:

 

I don't know how FT interprets his measures, so I can't comment on his use of CVD. However, I can issue a word of caution to you: If one blindly split buys and sells as is done in the industry today (two approaches used are: (a) using quotes to split the T&S data (the perils of which I listed in my earlier post to BlowFish); and (b) as was described in Orline Foster's work: "The Ticker Technique and the art of tape reading"), one will double count the split volume.

 

Once such double counts are rectified, CVD could be used as you mentioned (although I haven't had the opportunity to analyze it yet -- a project for later!)

 

Hope this answers your question.

[pmwhite 10]

@pmwhite, @gooni:

 

I concur with BlowFish's answer to your questions.

 

@gooni:

 

Further, BlowFish's comments and my response to his comment might provide you some clues as to why all market order buys at the ask are not necessarily "buys".

A small hint: By assigning all market order buys at ask as buys, one will be double counting buys! Enough said. As I said in my reply to BlowFish, BlowFish might be on to something!

Thanks to you and Blowfish for your responses. This brings me back to the reason I asked about the liquidity provider / taker definition. If I read your response correctly, then the order placement - namely limit orders determines providers and market orders determines takers.

 

However, with what you've posted in this reply to gooni and to Blowfish, it sounds as if a liquidity provider, who starts off getting hit on a limit order, and who then becomes a liquidity taker when said provider attempts to exit from the position/inventory using a market order in an effort to balance inventory, should not be treated as a liquidity taker but perhaps as a liquidity provider in spite of a market order exit. Is that the correct reading from these posts? Is this where supply/demand for liquidity providers comes in?

[madspeculator 10]

This brings me back to the reason I asked about the liquidity provider / taker definition. If I read your response correctly, then the order placement - namely limit orders determines providers and market orders determines takers.

 

According to my definition of liquidity providers and liquidity takers, yes, it is the placement of orders that determine the category.

 

However, with what you've posted in this reply to gooni and to Blowfish, it sounds as if a liquidity provider, who starts off getting hit on a limit order, and who then becomes a liquidity taker when said provider attempts to exit from the position/inventory using a market order in an effort to balance inventory, should not be treated as a liquidity taker but perhaps as a liquidity provider in spite of a market order exit. Is that the correct reading from these posts? Is this where supply/demand for liquidity providers comes in?

 

You are very close to the answer as per the proposed theory, but will have to "invert" your logic. Viewed within the framework of the proposed theory, the entity while accumulating inventory is a liquidity provider, and is a liquidity taker when disposing such acquired inventory. The reason for this line of thinking is to make the measuring process relatively easier compared to your line of thinking (which is also correct, but another way of looking at the same situation) makes measurements very difficult 'cos one has to look into the future to determine if the present trade is a "liquidity provider" trade or not (just my opinion).

 

Yes, this is where the liquidity providers supply and demand comes in.

[natedredd10 10]

For those interested in the tools I use to perform my research:

 

bravo, bravo...I barely know where to start in this discussion its so great.

I would like to first ask if you could do a more technical post on your software setup..This is very inline with the trading path I'm on but the biggest problem I've found is

a) retail trading software is shit

b) statistical analysis software like matlab and R are amazing because its not harder to learn than "trading scripting lanuguages".

c) since most retail traders don't use these packages though there is a HUGE interface problem with these packages to retail data vendors. Basically an API hurdle with your data provider of choice.

Are you actually interfacing to R via your data providers API?

While this is a great theoretical discussion, like urmablue's stuff on here...its vastly less usefull if you can't play with the actual toys yourself.

Edited January 29, 2010 by natedredd10

[natedredd10 10]

Everything is possible. However, I am yet to find proof of such value. Also, new research in finance and economics is questioning the existence of value as we presently understand it.

 

I do think its helpfull in a discussion such as this that we toss poisoned trading terms like S/R, "momentum" in the trash can..even "value" is a bit poisoned with bias on this board because of MP ideas.

Even "intrinsic value" is a poisoned term from too much economics literature...as if there is a machine locked away in Antwerp that knows the exact value of things and everyone is voting on what this machine has come up with. Its probly a heuristic hardwired in our brain to seek a leader such as this, even though none exists outside the illusion of the consensus in markets. The consensus does form sometimes though, which to me is what traders call S/R...the problem with that line of thought though is its like thinking the ocean's function is to create waves...the ocean is not trying to create waves, but waves are a property that emerges in a system like the ocean. Traditional ideas of S/R are poisoned by this bias that at that point exists some binomial distribution on what price is going to do above/below that point.

I don't see how you explain markets better than the concept of the minority game. It doesn't defy common sense, is not overly complex and exhibits exactly what we see in markets with the payoffs flowing to the minority.

With that in mind its hard to view myself as either a liquidity provider/taker..Part of the problem is the microstructure literature is not thinking about what we are doing as retail traders. The liquidity offered to me as a trader is basically infinite, the liquidty i offer is basically meaningless. 2 bit pirate surfing waves of liquidity, that the microstructure literature doesn't need to bother with.

[natedredd10 10]

However, the "false positive" (price going in the opposite direction than what was predicted), in my mind, is the true measure of risk for a trading model; my risk management is a function of my false positive analysis.

 

Totally agree that the absolute most poisoned trading term is predictability...

I really like the idea of "false positive"..False positives are why most people have a hard time strategically in trading.

My only problem though is that idea is too exact, too medical almost..

How do you control for the stochastic part of markets in your research?

[natedredd10 10]

This is exactly how client order execution algos work! You might be on to something here. Now, go one step further and think about how index arbs programs might work, and how index option market makers might trade index futuers (these two steps are necessary only if you trade index futures or futures that might be used for hedging other financial instruments), and you will get the "full" picture!

 

Index arbs are the elephant in the room in this discussion. The idea of thinking about index arbs inventory is senseless IMO because they may as well transact at the speed of light from our position...I don't see how it makes any sense to even bother trying to get a read on index arbs inventory as a liquidity provider...you will already be looking in the past the moment the data hits your screen.

I would love to know though what you think about index option market makers..I'm years away from fitting them into my optimal trading strategy but I would bet it all you can get a nice read on position sizing on index futures bets from the ability to read the volatility surface of index option market makers. Vastly better than something as absurd as ATR, or something you can even see inverse correlation with the naked eye by pulling up a chart of the VIX.

[madspeculator 10]

I would like to first ask if you could do a more technical post on your software setup..This is very inline with the trading path I'm on but the biggest problem I've found is

a) retail trading software is shit

b) statistical analysis software like matlab and R are amazing because its not harder to learn than "trading scripting lanuguages".

c) since most retail traders don't use these packages though there is a HUGE interface problem with these packages to retail data vendors. Basically an API hurdle with your data provider of choice.

Are you actually interfacing to R via your data providers API?

While this is a great theoretical discussion, like urmablue's stuff on here...its vastly less usefull if you can't play with the actual toys yourself.

 

I am not going to pretend that my infrastructure resembles that of a typical retail trader. I have a messaging bus the connects all components of my infrastructure. The components include a MySQL db, R, trading platform (home grown), and broker interface. I also have a CQG feed that I use for cross-checks.

[madspeculator 10]

Totally agree that the absolute most poisoned trading term is predictability...

I really like the idea of "false positive"..False positives are why most people have a hard time strategically in trading.

My only problem though is that idea is too exact, too medical almost..How do you control for the stochastic part of markets in your research?

 

I haven't found an acceptable way to control for "local" stochastic vol. The way I overcome the problem of estimating such "local" stochastic vol is by creating a proxy that is a function of the traded funds (my risk). The biggest problem with this approach is that I am using a proxy which is not correlated to the local stochastic vol. I am ok with it, for now.

 

If you or anyone else have any other suggestion, I am all ears!

[madspeculator 10]

Index arbs are the elephant in the room in this discussion. The idea of thinking about index arbs inventory is senseless IMO because they may as well transact at the speed of light from our position...I don't see how it makes any sense to even bother trying to get a read on index arbs inventory as a liquidity provider...you will already be looking in the past the moment the data hits your screen.

 

You are correct that the moves they produce are phenomenal, and so is their inventory effect. However, the reason one needs to be aware of those arb programs is to expect it. Also, you will be surprised that many a times, because of the way they operate, a trader who can recognize such trades will be able to get really good trade locations before the move starts.

 

I would love to know though what you think about index option market makers..I'm years away from fitting them into my optimal trading strategy but I would bet it all you can get a nice read on position sizing on index futures bets from the ability to read the volatility surface of index option market makers. Vastly better than something as absurd as ATR, or something you can even see inverse correlation with the naked eye by pulling up a chart of the VIX.

 

The usefulness of a vol-surface for a trader like me (primarily high-frequency) is very limited. They seldom change appreciably during the day unless there is an event, and they only show the option MM's collective long-term bias. It will be great if I could some how read each of the option MM's individual vol-surface (which will show their inventory bias), but that is not possible. On the other hand, may be, there is information in the vol-surface that a high-frequency trader can make use of, but I haven't discovered it yet.

Edited January 30, 2010 by madspeculator

[gooni 10]

OK so any single transaction is between, at that point in time, a liquidity provider (limit order) and a liquidity taker (market order). In order to differentiate whether to allocate said transactions as 'buys' or 'sells' is dependent on determining which side is expected to be holding inventory at the end of the transaction.

 

This I suppose is clear enough. I haven't yet deciphered the cryptic clues, especially the one about counting buys twice (although my thinking, so far, is that it's potentially counted once as a sell on the liquidity provider side then later as a buy on the liquidity taker side, which may have the 'effect' of counting it twice as a buy). Clear as mud? Is for me...

[BlowFish 10]

You are correct that the moves they produce are phenomenal, and so is their inventory effect. However, the reason one needs to be aware of those arb programs is to expect it. Also, you will be surprised that many a times, because of the way they operate, a trader who can recognize such trades will be able to get really good trade locations before the move starts.

 

 

Do you use other traditional market internal type indicators in the correlated market to try to help detect when these are going off e.g. Prem Tick Tikki etc.? I guess you must try to identify the arbitrage opportunity to get in with/ahead of these programs?

 

Incidentally arbs use the concept of 'value' but not in an MP sense, there assumption is the basis (difference in prices of the components of the hedge portfolio) will return to 'fair value'.

[gooni 10]

Incidentally arbs use the concept of 'value' but not in an MP sense, there assumption is the basis (difference in prices of the components of the hedge portfolio) will return to 'fair value'.

Yes, but there's a empirically measurable and practical idea of value, rather than some ephemeral concept that is often touted through MP. It is not so much value as efficiency.

[clmacdougall 11]

I do think its helpfull in a discussion such as this that we toss poisoned trading terms like S/R, "momentum" in the trash can..even "value" is a bit poisoned with bias on this board because of MP ideas.

Even "intrinsic value" is a poisoned term from too much economics literature...as if there is a machine locked away in Antwerp that knows the exact value of things and everyone is voting on what this machine has come up with. Its probly a heuristic hardwired in our brain to seek a leader such as this, even though none exists outside the illusion of the consensus in markets. The consensus does form sometimes though, which to me is what traders call S/R...the problem with that line of thought though is its like thinking the ocean's function is to create waves...the ocean is not trying to create waves, but waves are a property that emerges in a system like the ocean. Traditional ideas of S/R are poisoned by this bias that at that point exists some binomial distribution on what price is going to do above/below that point.

I don't see how you explain markets better than the concept of the minority game. It doesn't defy common sense, is not overly complex and exhibits exactly what we see in markets with the payoffs flowing to the minority.

With that in mind its hard to view myself as either a liquidity provider/taker..Part of the problem is the microstructure literature is not thinking about what we are doing as retail traders. The liquidity offered to me as a trader is basically infinite, the liquidty i offer is basically meaningless. 2 bit pirate surfing waves of liquidity, that the microstructure literature doesn't need to bother with.

 

Thank you for the post, I really enjoyed reading it.

I've begun to see the markets as a "Trapping Structure" perhaps that's just a different way of putting what you have posted as a "minority game". All the same I am learning to identify the trap being set by the hunters out there, it's enjoyable to see that their actions are repetitive even if the market isn't.

[FulcrumTrader 10]

I also have a CQG feed that I use for cross-checks.

Hey Madspeculator, I see you have CQG....I just set myself up with another futures account to try out CQG (because they have excellent bid/ask data) and eventually see if they could add Cumulative Delta candlsticks capability. They already have the data formated properly for their TFlow charts so it should be VERY EASY for CGQ to add CD candlesticks. I would LOVE to one day have the charting/trade platform/and data source all combined for my Cumulative Delta charting/trading needs.

[BlowFish 10]

BlowFish:

 

This is exactly how client order execution algos work! You might be on to something here. Now, go one step further and think about how index arbs programs might work, and how index option market makers might trade index futuers (these two steps are necessary only if you trade index futures or futures that might be used for hedging other financial instruments), and you will get the "full" picture!

 

 

I guess I could have rolled this into the previous post but separate train of thought. Trouble with arbs is there are so many types. For this I guess it's Ok to leave aside the esoteric types and focus on the S&P. I guess that means we are looking at the cash as the reference instrument? That would make the futures and or the options the hedge. Lets leave aside options for now. I find it hard enough to think about 2 things at once let alone 3!

 

Can we consider arbs in a similar way to the traders we have discussed?....heres my thinking. Arbs will essentially try to provide liquidity in the markets they trade as they are legging in and out of positions. The cash position (being the reference instrument) may leave our Arb requiring inventory for there hedge. Like the trader executing for the elephant they might need to take liquidity form the hedging market (the futures) to balance out there hedging portfolio. Does this sound plausible?

 

I think I can see the pieces of the puzzle sitting there on the table but I am not sure I am closer to putting them together! (i.e detecting things) Might be time to start drawing boxes and arrows.

[Karish 10]

Pardon for my lack of knowledge, can you please explain why during the process of distributing large stock block, the algo provider need to buy inventory. Even if they do need to buy some inventory during the distribution process in order to minimize their effect on the stock price, one would expect that the algo process would clear the unnecessary inventory along the process, and not to carry to much to the end.

 

Thanks, Karish

 

These algorithms, in the process of filling a trader’s order, end up buying or selling more quantity that is really needed. This excess quantity, in the hands of the algo-trading providers, then becomes “inventory”.

[BlowFish 10]

Pardon for my lack of knowledge, can you please explain why during the process of distributing large stock block, the algo provider need to buy inventory.

 

 

 

 

This is exactly how client order execution algos work!

 

My emphasis. Fulfilling the client order may leave you holding some 'working' inventory or even taking some of the clients inventory just to finish their order.

[Karish 10]

Thx BF,

 

One of the theory's tenants is that during weak order flow that liquidity provider will trigger Stop Loss, any idea if we should categorize it as part of their inventory balancing or it is just making easy money?

[djangho 10]

The obvious answer (well to me so it might be half baked) is that they would try to offset the position to other side liquidity takers. If that is not possible they would demand inventory from other side liquidity providers. I guess one might assume the very fact they have ended up with an inventory imbalance would mean they might swiftly move to step two, demanding liquidity?

 

Isn't this a fundamental issue when there is an intermediary the intent liquidity requirements of the original participant is obscured?

 

How can they demand inventory from other side liquidity provider?

[BlowFish 10]

How can they demand inventory from other side liquidity provider?

 

By hitting the bid/ask.

[AgeKay 10]

Thanks for presenting your theory. Very interesting. I was able to follow your abstract description but when I came to the following part I was wondering how something like this would look like an actual trade scenario:

 

The author’s own work presented below tries to estimate the demand and supply of both the liquidity providers and liquidity takers. By monitoring the demand and supply of liquidity providers, the author is able to notice the inventory adjustment process [...] NIP measures the inventory held by liquidity providers, and NOP measures imbalance in the liquidity takers’ order flow. The sum of these two parameters is the “real” order flow imbalance.

 

A) Could you give an example on how one would determine liquidity providers and liquidity takers with T&S and/or quote data? Something like market goes bid, it's bid so many contracts, trades so many contracts, then trades the offer until its cleared, offer goes bid, etc. Who would be the liquidity provider vs. the liquidity taker?

 

B) And how do you determine when liquidity providers are balancing their inventory (I am referring to tenant 4d)

 

C) How are liquidity providers reacting to the increased volatility of the past days? They can't balance their inventory with so much order flow and so little liquidity when markets are so volatile.

[djangho 10]

question removed, sorry for that

Edited May 8, 2010 by djangho