Foreword

Financial Market Theory (General)
For many years financial market theory has been synonomous with Capital Market Theory (CAPM). CAPM is based on a gaussian distribution. In times of balance or gentle market movement the gaussian applies. In strong markets, it does not. The most marked shortcoming is in the underestimation of risk. Recent work by econotheorists indicates that not only does the gaussian distribution not adequately describe financial markets, but it is unlikely that any distribution can be found that will. This article accepts that financial markets are complex, self regulating and driven by feedback. We propose a theoretical framework for understanding markets as well as methodology that decodes the feedback, permitting a trader or analyst to discern value and value change. The lack of a characterizing distribution function makes market predictions or calculating long term risk difficult, if not impossible. The theory presented here is from empirical studies that provide logical, tested and well supported explanations for a number of observed facts. The underlying principle is that value can be determined from market feedback.

The value approach treats market analysis as a science, as defined by the Compact Oxford English Dictionary: "Science is the intellectual and practical activity encompassing the systematic study of the structure and behaviour of the physical and natural world through observation and experiment." The basic observable is market feedback, principally seen as reports of market activity in the form of ticks and other short term data. The foundation of markets is value and value change as measured from market feedback. A market's fundamental variables are price and volume (often from streaming tick data), value from feedback and the feedback relaxation time, the time it takes for a change to propagate through the system.

Financial Market Theory (Financial Economics)
Modern financial market theory began with the studies of Markowitz in the 1950's. Sharpe completed the work in 1970 with the hallmark book, Portfolio Theory and Capital Markets (1). Capital Market Theory, CAPM, developed in the book, assumes equilibrium markets and a gaussian distribution of price range, with decisions being made by a 'rational investor'. CAPM defines market risk as the standard deviation of the distribution. A basic assumption is that investors consider only expected return and risk, an assumption shown to be incorrect by behavioral economists.

Behavioral Economics (Ref 2)
The psychology of decision making shows that humans deviate widely from the standard economic model. 'Bounded Rationality' considers the limited cognitive abilities of investors; 'Bounded Willpower' shows that people may make choices not in their long-run best interest and 'Bounded Self-Interest' involves investors sacrificing their own interests to help others (e.g. charitable giving).

Economic Fundamentalism This is an implicitly linear way of evaluating markets. Each market variable is identified and ranked. In the Dow Index one can understand that the employment picture affects the value of market instruments (stocks, futures, options, etc.). So also does taxation, the balance of trade, exchange rates and on and on. There are some forty government economics reports that in some way describe the economy and hence affect the market value of stocks, and these are not the only 'fundamentals'. It would seem that a simple commodity like silver, would be easier to understand. Production is known and the major uses are relatively few (jewelry, solid state devices, etc.). But no matter. The market can move in contradistinction to all known information (witness the silver market in the 1970's, when price moved from a few dollars per ounce to nearly $50--all due to the unknown and unknowable effect of someone seeking to corner the market). Fundamentalism fell into decline with the advent of home computing, giving way to technical analysis and charting in the late 1970's and 1980's.

Mathematical Fundamentalism
Mathematical fundamentalism is a logical extension of economic fundamentalism, considering each fundamental element as a variable. Here one takes the list of all the known elements that can affect a commodity and develops an encompassing equation, e.g. if there is a shortfall of the soybean crop of 10 percent, price will rise 20 percent. But what if the Argentine crop is exceptionally large? The many variables interact. Some interactions can be evaluated, some cannot. Some factors are totally unknown, and additionally they interact in complicated and non-linear ways. Assuming the sorts of problems above can be solved or estimated, a mathematical description of a market would be a non-linear, non-homogeneous differential equation with non-constant coefficients. Not knowing all the interactions between the many variables means that such a market equation could not be written nor solved for real world markets. A mathematically correct fundamentalist solution of market behavior is not possible.

Econophysics (EP)
The econophysicists entered the scene in the mid 1990's, focussing on the market distribution function. EP shows that markets are often not in equilibrium and that the gaussian distribution is unable to describe the rich texture of auction markets, particularly the "fat tails" of distributions that can and do exhibit inordinate risk. EP finds that markets are complex systems, systems that are in continual change from the feedback of the market's agents (traders). Like CAPM, EP relies on the observables price and price change. EP recognizes the need for a valid mathematical model or distribution function for describing markets; but concludes that such a distribution function may not exist.

Value Market Anaysis
Evaluating markets from the value standpoint was initiated in 1985 when J. Peter Steidlmayer, at the Chicago Board of Trade (CBOT), presented the Market Profile (MP) as a market informational tool (9). Data for the MP is the (CBOT) cleared volume and trading activity. The aim of MP initially was to show the structure of the trading and the value (range) at the end of day. Later, CBOT began to announce clearings during the day; but the cleared information always lagged the market, so MP and value remained useful primarily at end of day (EOD). The EOD knowledge of market behavior and value gives the trader an an edge for next day trading. MP differs from CAPM and the later EP concepts in that the primary variable is value, although the primary observables remain price and price change. A subsequent text by Steidlmayer and Koy (10), explored the inner workings of auction markets.

Extension of the Original Value Analysis
The Steidlmayer & Koy book (ref 10) and the CBOT Manual (ref 9) which preceded it, dealt with CBOT futures exclusively. This limitation was overcome two years later with the Meta-Profile (ref 15). This new kind of profile used tick data instead of the cleared volume of the CBOT profile. The Tick-TPO or Meta-Profile, could be calculated for all markets that produced tick data. Jones then took the profile into multi-day analyses (ref 7, 8). Professor Drinka at Western Illinois U. explored the details of the profile and initiated the Market Profile Society International trade organization. Then in 1993 (ref 11) Jones developed analyses for evaluating the effects of commercial (insider) traders, including finding an essentially zero day-to-day serial correlation in futures markets. Day-to-day volatility studies by Jones showed large changes over short times (2003 (13)). Additionally, a number of Jones' other publications addressed nuts and bolts issues of actual trading, ending with an early version of Auction Market Theory in 2002 (14).

Auction Market Value Theory (AMVT)
The market is accepted as a complex system, self regulating and driven by feedback. Value is the primary variable for describing the market. Subsidiary variables are price, time and volume of one sort or another. In place of an overarching distribution function, AMVT examines the many component parts of a market (reference points). Each component describes an aspect of the market. The market is a puzzle; the component parts are the pieces of the puzzle that must be fitted together. The result is a description of the market (value, market condition, volatility, volume, etc.).

The sum of the descriptors, collectively, describe the market as a whole. Thus, the goal of AMVT is a clarification of the auction process, using the component parts of market to gain understanding entire entity. A premise of Auction Market Value Theory is that an overarching mathematical formulation of a market distribution is not possible with current market understanding and mathematical capabilities . Indeed, in view of the complex nature of auction markets, a closed form solution may never be possible. The utility of AMVT is its potential for explaining the current state of the market.

At bottom, a theory should be an aid or pathway for decision making. One would like to predict the future to some degree. In markets, the complexity precludes true forecasting. However, AMVT provides information reliable enough so that market decisions can be based on the AMVT data.

Contents of Auction Market Value Theory
1.    Theories
2.    Market Profile Defined
3.    Overlay Demand Curve Defined
4.    Auction Market Value Theory
5.    Auction Market Knowledge:
6.    Auction Market Knowledge: The Longer Timeframe
7.    Development of an Overlay:
8.    Market Condition from Overlays:
9.    Calculating Risk
10.    Auction Market Value Theory Reviewed
11.    Applications
12.    Short Covering Rally
13.    Buy/Sell Confirmation of the Original Premise for Short Covering
14.    Commercial Capping
15.    Volatility
16.    Value Areas from LDB and Market Profile
17.    Conclusion
18.    Unfinished Business

Section 1
A Brief Review of Current Market Theories

CAPM: An Equilibrium Theory of Financial Markets
Many commentators have criticized CAPM's assumption of equilibrium. The observable, daily high-low ranges, often appear to not be in equilibrium. Volatility is defined as the standard deviation of the price - time distribution and it is also defined as risk. This too, drew comment since opportunity is present and must somehow be related to price behavior. Actually, the very existence of a standard deviation requires a known distribution. The distribution assumed, gaussian, is demonstrably not describing the market much of the time, hence the standard deviation is devoid of meaning at those times when the market is not in equilibrium. Most certainly, if the standard deviation is not defined, the volatility cannot be either.

Ultimately, CAPM came to be understood as pertaining to well behaved (equilibrium) markets. But it could not explain drawdowns, crashes and other extreme behavior emanating from non-equilibrium activity. Furthermore, as the Behavioral Economists have discovered, the supposed 'rational investor' may often act in irrational ways.

The Normal distribution of the daily trading range is well understood, where the first moment is the mean and the second moment is the standard deviation. Analytical advances can be made with this distribution that would be impossible if the distribution were more complicated or not known at all. An example of such an offshoot is the Black Scholes model for option valuation. A contributing factor in accepting the normal distribution for markets was the then prevalent Efficient Market Theory. If the market is efficient, discounting all new information instantaneously, no one could find an edge for trading profitably.

Although there has been much negative comment on the equilibrium market proposition, the econophysicists have done the best job of showing why the stock market is not stochastic (JJH 34, 63-64).

Aside from the stochastic problem, the Sharpe book contains a theory with valuable insights for portfolio management and the afore mentioned options valuation. By and large CAPM does not describe real markets (they are not stochastic) and option valuation founders on the definition of volatility. Although CAPM is useful in balancing portfolios, it offers little help on the mechanics of buying and selling. If the market were truly stochastic and efficient, it should make no difference when the new stocks are bought and the old sold. But the market has been shown to not be efficient; so, theoretically at least, timing, denied by efficient market theory, can possibly be profitable. In the absence of a stochastic process, i.e. when the actual market is not in equilibrium, proofs based on stochasticity are invalid.

The fundamental problem of defining annualized volatility as the standard deviation of the annualized market has led options traders to seek a more current volatility. Implied volatility is backed out from the actual market pricing of options. A problem with implied volatility is that the iterative mathematical procedure involved is captive to market noise, particularly since price is the primary variable. Further, if the market's distribution is not gaussian, as is shown by the Econophysists, higher moments are significant in evaluating risk, i.e. the simple case cannot find dangers present in the real world. Implied volatility is reputedly a significant culprit in the spectacular failure of the trading firm Long Term Capital Markets ($1.3 trillion loss) (RL).

While it is now accepted that the stock market is demonstrably not a stochastic process and volatility is demonstrably not just risk, the real difficulty with CAPM is that markets are a complex process and CAPM is based on the exceedingly simple gaussian distribution and rational investors. At times the market is well behaved and the gaussian model is adequate. At other times, when a gaussian is not appropriate, the dangers may be extreme. One has a devils choice: use a simplified picture (equilibrium) and be able to analyze the market (knowing there will be times when the analysis is wrong) or admit the market is complex and be able to analyze nothing within the distribution format. The decision is not quite so draconian today. The field of Econophysics, addressing markets as complex systems, is making advances toward a market analysis that includes elements of reality, such as crashes. Further, the reality approach of Auction Market Value Theory offers a manner of dealing with complex markets based on an understanding of the principal elements of markets and how their reaction to feedback can be understood.

In spite of the substantial and known drawbacks with CAPM, the theory continues to be taught in financial MBA programs. This is probably a result of there being no logical successor theory and CAPM is teachable. By "teachable" we mean:
1. It is coherent and complete
2. Mathematics are sophisticated enough to make MBA level students
     work on the mechanics.
3. Tests can be made exact and thus easily gradeable.
4. The general ideas can give students a grasp of capital markets.
5. MBA students who are not planning to be traders (the majority).
     thus have enough information to get by.
6. Portfolio balancing/efficient frontier and other portfolio
     concepts are impressive and useful. 7. Mathematically oriented students can manipulate CAPM to gain
     new uses, e.g. Black Scholes option pricing


Econophysics is attempting to understand risk, to explain extreme events like crashes and basically to explain why markets behave as they do. Like CAPM, EP takes as observables, price and return (price change).

Modern Market Analysis, Complex Systems and Econophysics
Recent work by Johnson, Jeffries and Hui (JJH) in the emerging field of Econophysics treats the financial markets as complex systems. They consider the problems of:
    1) Markets as complicated dynamical systems that are continually
    generating high-frequency data series.
    2) How the stochastic assumption gives misleading answers to practical
    problems such as minimizing risk, explaining extreme events such as
    drawdowns and crashes and pricing derivatives.
    3) Why financial markets behave as they do.
    4) What can be done to minimize risk.

JJH on page 2, lists as goals, these practical questions:
1) When to buy.
2) When to sell.
3) Risk.
4) Predictability.
5) Crowd behavior.
6) Forecasting on basis of crowd behavior.
7) Forecasting time evolution of markets.

Complex systems concepts have only recently begun to be applied to financial market analysis. Complex systems are generally nonlinear with feedback acting to continually adjust the system. There appears to be no hard and fast definition of a complex system.

Peters in 1999 (EEP) listed these characteristics of a complex financial system:
1) The system has a purpose (e.g. to facilitate trading of say, soybean futures, stock market indexes, etc.)
2) The system is decentralized (many independent agents/traders)
3) Feedback occurs within the system (all agents observe the system and make changes in their behavior)
4) The system adapts to information from feedback (losers get out, winners increase holdings)
5) Adaptations are decentralized leading to innovation (each agent makes its own decisions)
6) Rules govern the system, rules can change or be changed (adapt) e.g. the movement from exchange floor to computer trading

Observations 1 - 6 describe auction markets such as stocks, interest rates, futures, derivatives and actually even markets as diverse as food and department stores (Ref 10). The financial markets are double sided auctions where a buyer at one moment may become a seller the next moment.

Items 2 - 5 show that feedback is diverse, affecting each agent or trader in a unique, personal way. The net effect of the feedback manifests itself in macrosocpic market parameters such as price movement and volume. But it is impossible and not even desirable to isolate a particular feedback element from any one of the agents; the useful observable is the net market change related to that element.

JJH defines complexity in financial markets somewhat differently from Peters.
1) Feedback: change contains an element of remembering.
2) Non-stationarity: the statistical distribution changes.
3) Many agents: traders, institutions interact in time-dependent ways.
4) Adaptation: agents adapt their behavior to improve their chances.
5) Evolution: agents behavior evolves thru feedback and adaptation and the system may not be in equilibrium. It can exhibit extreme behavior such as crashes.
6) Single realization: i.e. averages over time are not equal to averages over ensembles.
7) The market is an open system coupled to the environment: one cannot discriminate between exogenous (outside) and endogenous (inside) influences.

JJH's analysis turns on measurement of price and return (JJH 16).
Briefly, they assume:
1) Price as a function of time is the primary observable.
2) The assumptions of CAPM that price changes are independent and identically distributed are not borne out by observation.
3) CAPM volatility is not sufficient to classify risk.
4) A market price series' expected value depends in part on previous movement, the market has some memory. This helps explain drawdowns and crashes.
5) Crowd action plays a role in volatility.
6) The 'zero risk' in writing options under CAPM assumptions is not true in real markets.
7) Real price series differ from the random-walk model.
8) Trader's beliefs/actions can create patterns sometimes leading to crashes.
9) Markets are non-linear.
10) Internally and at all times, price moves much more and faster than rational expectations (of return) would predict.

The point is that both econophysists and Peters see the markets as complex systems with feedback causing change dynamically. The econophysicists are more mathematically oriented and propose a stricter analysis. They also propose a mode of attack, one using distribution functions. At this point it is not clear how a general market distribution function might be developed. Looking at the complex market from a differential equation standpoint, a descriptive equation would likely be nonlinear (feedback probably does not behave linearly), non-homogenous (variables most likely could not be separated, even if they could be defined) and the coefficients are non constant (an increase of ten percent in the soybean crop would rarely translate into a change in price of ten percent). It would be most unlikely that a complicated differential equation could be translated into a distribution function. One might think of a complex market in terms of the parable of the blind men and the elephant. An analyst explores all the various parts, ending up with enough pieces to make a slow moving elephant. However, with a complex market, the feedback can drive a change so that the next time the various parts end up making a fast moving tiger.


Behavioral Economics (BE) revolves around the fact that humans have limited brainpower and sometimes cannot reach the optimal decision; we often do not exercise the proper willpower and we sometimes (often) act unselfishly (about 75 percent give to charity). Aside from the personal foibles of lack of will power and unselfishness, we cannot be certain that we are making the best investment decisions. In fact, in a complex market, the feedback, the data presented by the market can never be totally deciphered. Investors have opinions. Sometimes the opinions are correct, sometimes not. The complex nature of the market insures that the investor will rarely, if ever, "be sure". Thus, the portfolio theory dictum that investors consider only expected returns and risk as measured by the standard deviation of returns is not evident in real life.

The Current State of Financial Market Theory: Real Markets
The real markets people trade are far from the idealized CAPM. The non-stationarity and feedback aspects argue strongly for markets to be non-linear, at least at times of high risk. Markets are not efficient, rather they are effective. The standard deviation of the annual market measured on a market day basis not just risk. It also includes elements of opportunity. Curiously, little of this information was addressed in the foreword of the 2000 reissue of CAPM.

Theory as it applies to Auction Markets
Theories are often thought to be incomplete without the ability to make predictions. This is not so. At base, a theory is an explanation of a phenomenon. A thorough understanding of the phenomen may well lead to the ability to predict at some level. However, the feedback nature of a complex market may preclude an ability to develop lasting predictions. For example, if the theory were found to predict tomorrow's price, everyone would soon learn and the feedback would adjust the market to make the prediction fail. This is obvious in futures markets which are essentially zero sum (in the absence of transaction costs).

Predictions that do not directly bear on profiting from trading are not disallowed. A study of the persistence of trends found that a mature market experiencing a new high price had an eighty percent probability of reaching a yet higher high within ten days (DJ1). This finding illustrated that the market had momentum under certain conditions, a feature not allowed in efficient markets.

The complex nature of the market is a strong argument for a phenomenological approach to market theory. A market has many facets, elements that are functional parts, but still with a large degree of independence in their behavior. As one of these elements changes, that changed behavior is returned to the market via the feedback mechanism. While elements such as volume, volatility, range, time behavior and the like are individually measureable, they all combine to form the market one studies or trades. Each element must be discovered and studied on its own before there is any thought of considering how it affects the whole. The vital question of extreme market behavior such as large, fast changes in value, must be understood. It is the thesis of Auction Market Value Theory that each of the market elements can be isolated and must be understood prior to forming a coherent description of a market. The experienced trader collects all the market information available, assesses each piece, ending with an understanding of the market situation. Not all pieces are necessarily needed to find the market's condition. Different sets will be used at different times.

As an analogy, think of a homicide detective. Experience has taught that there is a wide variety of clues. Arriving at a crime scene, the detective collects all clues available. Many clues are irrelevant and are later discarded. The pertinent clues lead to a solution of the crime. At the next crime scene, again all clues are collected. Again, the irrelevant clues are discarded, except this time it is a different set of clues that are irrelevant. If you were to suggest to the detective that to save time only the relative clues be collected, you would be laughed at: which clues are relative becomes known only through the discovery process of the crime investigation.

A crime scene is a complex situation in real life. The perpetrator seeks concealment, hiding or leaving misleading information. It is the detective's job to uncover the truth. The real world of the auction market is merely another a complex process with many clues of varying utility at various times.

Auction theory has the goal of understanding the current market situation. Is the market balancing, trending or behaving in some other way? Obviously if one can understand the current market condition it was possible to understand the condition at an earlier time. Comparing the current with the earlier condition allows one to see if the market has changed.

A combination of the relevent market elements will lead to the conclusion that the market is behaving essentially as it was at the prior determination or it has changed. No change means continuation. If the market was trending the trend continues. Or if the market was congesting, it still is.

Change or continuation provides the basis for market decision. For instance, a trader can find opportunity in change. A conclusion that the market is changing out of a balance condition may lead to a decision to trade. A change from moving (trend) into balance is an argument to exit from a trade. In the case of a portfolio update the old equities should be eliminated in a balance or change to downtrend situation. The new equities should be bought into the portfolio when they are in balance or trending up. If they are in a downtrend, it clearly makes sense to wait until the downward move is over before purchasing them.

To the analyst/trader it all boils down to a determination of change or continuation. Another way to state this is to couch it as a search for the unexpected. If a market has been in equilibrium, it is expected that it will continue in equilibrium. If a market is trending, it is expected to continue. Change is the unexpected. One way or the other, an analyst/trader must search for the unexpected. The expected, the market condition (e.g. balancing or in equilibrium) must be known. The only way to find a market's condition is through empirical measurements. Sharpe, in Ref 1, pg viii, decries empirical studies, stating they have a "short half-life". In the absence of an over-arching theory like the normal distribution of CAPM, complex market analysis must rely on empirical market measurements.



What Drives Markets
Fundamentally, markets are the place where buyers and sellers meet to exchange goods. Early on the commodity grain markets were initiated as a place to lay off risk. In December, a baker who needed 5000 bushels of wheat in July could buy a contract at the exchange for delivery at that time. The seller contracted to deliver the wheat at the appointed price at the appointed time. The baker had price protection for the bread he had to deliver seven months hence. But market conditions may change over seven months. Enter the trader, the speculator, who attempted to take advantage of the changed situation by buying or selling. Value of the wheat would most likely change over time and so the market offered opportunity to the trader.

Just as the Behavioral Economists are reality driven, even more so is the market. Market prices vary. This offers an apparent opportunity to the trader who is not a user of the commodity, rather one who seeks to profit on price change. The precepts of Auction Market Value Theory provide ways to read and understand the market and so the intelligent speculator is armed in the way that any other professional is armed by knowledge of his/her field. An external element, not a part of the theory is the background business climate in which the specualtor acts.

From the institutional (brokerage, exchange member) standpoint auction markets are fee driven, not profit driven. Win or lose, the broker gets a fee. Investors buy and hold, minimizing fees. The speculator makes many trades and generates large fees. A discussion of what drives short timeframe trading in markets might be illuminated by stating that brokers' success in advertising for new speculators is a major driving force. A fund of new money continuously keeps markets lubricated. Other trading, e.g. by hedgers and, in equities, portfolio roll-over is a source of liquidity exclusive of speculators.

An institutional fee driven mindset is not unreasonable in view ot the statistics. An estimated ninety percent of speculators will disappear within the first year they are active. A solid broker-trader relationship is hard to develop when the customers keep leaving. If indeed the trading business is not speculator friendly, then one element for success and a defacto part of auction theory is for the speculator to be prepared. In other professions a sometimes extended training period is normal. Doctors, lawyers, engineers, architects all undergo years of study. It is an adjunct of auction theory that anyone trading will better their chances by understanding their career field.

Gain from investing/trading in markets is the driving force in market research. The questions are simple: where should I enter, where should I exit, how can I minimize my risk between entry and exit and how can I define my risk? Secondarily, what information can I obtain to help find answers to my trading questions? Lastly, will my methodology aid in predicting the future course of the market? These questions can be characterized as all trading related. Reduced to the minimum, question 1 is "where to get in" and question 2 is "where to get out". Question 3, prediction of future events, is principally negative--predictions of complex processes are highly limited.

CAPM, for all it's benefits to portfolio operations, is silent on the trading aspects. Indeed, if the market is a random walk there are no preferred entry or exit points. Risk then is the standard deviation of a year's daily trading ranges. The market demands an entry and exit point. The trader must specify the entry/exit price in the order to the broker to initiate a trade.

Econophysics attacks market understanding from the distribution standpoint. How reliable is the assumed gaussian distribution? Where does it fail? What are the consequences of failure? Are deviations from the Normal curve significant in real world terms? Real markets are complex, so the higher moments of the distribution (whatever it really is (quasi-gaussian or other)) are significant. Econophysics has made real advances in critiquing CAPM. They still have far to go to reach the goals stated in JJH: when to buy and sell, defining risk, understanding crowd behavior and forecasting.

Auction Market Value Theory seeks to uncover the root causes of what drives the market in question, not all markets equally since each market at a given time will be in some state in its cycle (balance, trend, balance). A market in balance is acted upon different market forces than a market in a trend. A trending market may be picking up steam or slowing. A balancing market may be dynamic (becoming more active) or reactive (slowing). Many variables are involved, each with a particular importance to that situation. Some of the variables are value, volume, volatility, trading range, cash flow, outside information (a Fed announcement on interest rates) and so on.


Trading in the Real World
Far removed from the halls of ivy, real traders (RTs) trade real markets successfully. These are people who have no idea of what the moments of a distribution function might be, yet they daily achieve the goals, the practical questions posed by JJH. They find when to buy, when to sell, how much risk to take and sometimes how to predict the market. How do they do it?

It is not enough to pass off the winners, those at the tail of the win/loss distribution, as there by chance. Many RTs are in the winners circle regularly, year after year. Trading is their profession.

How Do They Do It?
No doubt some regular winners trade by 'feel'. Others use market analyses and charts to guide them. However they do it they are solving the market equation in real time. It appears that most of these regular winners put together a deep understanding of market behavior with key features of market action being the tip-off to buy or sell or exit.

JJH speaks of 'chartists' and 'fundamentalists'. Chartists supposedly predict the direction of markets from chart patterns while fundamentalists try to find the 'right price' from supply and demand considerations. Another methodology for analyzing markets, called technical analysis, became popular in the early 1970's. A principal tool of technical analysis is the moving average of prices, where an average of, say 20 days is used in place of the current price. This smooths out much of the volatility noise while moving the center of gravity of the average back to the middle. A shorter average, say 10 days, will respond to change faster than the longer and traders take their cue from the relative changes. Technical analysis tends to work best in well-behaved, slowly changing markets, just the kind that CAPM works well with. But, as with CAPM, fast changing markets are a problem for technical analysis. RTs may use all three methodologies and doubtless some of their own devising. However, the techniques the consistent winners use can only be guessed, since only a fool would divulge a winning methodology (which would quickly become a losing methodology because of feedback).

An unlikely event in practical market analysis occurred in 1985. One of those consistently winning RTs, J. Peter Steidlmayer, a member of the Chicago Board of Trade, revealed that his trading basis was value, not price (Ref 10). But his value is not the value sought by the fundamentalists. Steidlmayer's value is found by calculating price over time. Daily, a market will have a high and low price which is traded little (auction markets reject extremes). Prices between the extremes are traded more, with maximum volume nearer the central prices. Steidlmayer named the price - volume diagram the Market Profile. A Market Profile graphic displays price vertically and volume on the horizontal. Value is the heavily traded middle of the profile distribution.

Section 2
Auction Market Value Theory The CBOT Market Profile (tm, CBOT)

A Market Profile is a graphic displaying price and volume on the vertical axis, with cleared price activity on the horizontal broken down by trading period. Market Profile is a subset of the Liquidity Data Bank (LDB) report. LDB reports are released only by the Chicago Board of Trade (CBOT). The horizontal activity is identified by letters for each time period (for half-hour periods the letters are A = 08:00 to 08:30, B = 08:30 to 09:00 and so on). These letters are called BRACKETS or TPOs (Time-Price-Opportunity). Value is defined as those prices included within the central 70 percent of the volume, beginning at the peak volume of the day. Market Profiles are available only from exchanges that report volume at price. Market Profiles come from cleared data and hence are not available in real time.

The CISCO Meta-Profile
A Meta-Profile replaces the volume of the Market Profile with tick data, generating TPOs (That-Price-Occured or Ticked). These TPOs are used as a surrogate for the volume of the CBOT Market Profile. Value is defined as those prices included within the central 70 percent of the TPOs. Meta-Profiles are generated in real time and are limited only to markets that produce tick data. Meta-Profile methodology was created by CISCO in 1987, it was published in 1987 and it has been in continuous use on the CISCO Bulletin Board and website since then. In balanced markets values from Meta-Profiles and Market Profiles agree quite well. In directional markets they diverge, an activity quite useful to traders and market analysts.

       DOW JONES Future, Dec 2003 Delivery
         20030915   07:20: -  15:15 CST
  
            Volume          Value
    -----------------------------  
     94550  $B                 
     94540  $B                 
     94520  $B                 
     94500  Z$ABC              
     94480  Z$ABC              
     94460  Z$ABC              
     94440 *Z$ABCDLM           
     94420  Z$ABCDLM           
     94400  $ABCDIKLM         |
     94380  BCDIKLM           |
     94360  BCDIKLMN          |
     94340  BCDIKLMN          |
     94320  BCDIKLMN          |
     94300  BCDEFHIJKLMN      >
     94280  BCEFHIJKLMN       |
     94260  BCEFHIJKLMN       |
     94240  BEFGHIJKLMNO      |
     94220  BEFGHIJKLMNO      |
     94200  BEFGHIJKLMNO*     |
     94180  BEFGJKLNO         |
     94160  BEFGJKNO          |
     94140  EFGO              |
     94120  EFGO               
     94100  EFG                
     94080  EG                 
     94060  EG                 
     94040  G                  
     94020  G                  
     94010  G                  

  Figure FOR1. Meta-Profile for the Dow Jones index future for
  2003, September 15.  Time covered: 07:20 to 15:15 central time.
  Column 1 is descending price, column 2 (the letters) is volume and 
  column 3 (vertical bars) identifies, value, the more actively traded region. 
  Prices near the high show low volume, as do prices near the lows. The 
  vertical bars encompass the central 70 percent of the volume, a region 
  named the 'value area'.
 

Section 3
Overlay Demand Curve for Market Condition

A longer term auction market construct is the Overlay Demand Curve, a collection of profiles summed linearly. This artifice is similar to integration since the day-to-day randomness in the profiles tends to cancel out. The resulting display is similar to figure FOR1, just covering a longer time. An advantage of the longer timeframe is the ability to view an arbitrarily long market run. The form of the curve gives information on balance periods, trending periods and periods dominated simply by noise. A period's identification, be it balancing trending or something between, is called the 'market condition'.

Two other types of auction market information are the Liquidity Data Bank (LDB) and the BuySell data. In addition to a profile like figure FOR1, LDB data posts the trading by type of exchange member. LDB shows how much volume was created by which type of member at each price traded within the day. BuySell data further breaks down the LDB by how much of the volume was buying and how much selling by each type of member at each price.

Section 4
Auction Market Value Theory