Menu
Sister Sites

Options

Bookmark Thread

First Page First Unread Last Page Last Post

Print Thread

Similar Threads

Auction Market Theory and Market Profile 25,556 replies

MarketProfile/Volume Profile/Auction Market Theory indi? 17 replies

ECN && STP && Scalping && Hedging - Broker Review - Oct 2009 24 replies

Auction Market Value Analytics Charts & Data 14 replies

Analytics, recommendations - S&P 500, Euro and other 11 replies

Attachments: Auction Market Value Theory & Analytics
Exit Attachments

Auction Market Value Theory & Analytics

Predictions from Charts, A Critique
Tested with Real Data

Chat at IOAMT: August 25, 2011

Donald L. Jones
CISCO Futures
August 25, 2011

Analysis of auction markets is complicated because of the complexity of the markets themselves. There are many agents (traders), each with agendas of their own. There is constant change in how the players view the market. Market fundamentals are in a constant flux, much of it beyond any trader's ability to learn. Trading is a competition, each tries to win to the disadvantage of the others. Many trade on the news, on calculations (technical analysis), on hearsay, on charts, on tarot cards, etc. But the markets react in ways driven by the wide variety of market descriptors (reference points) present. Naturally, traders seek an 'easy' guide to trade selection. Charts are 'easy'. We will show in this talk that charts can mislead and better yet, there are valid, verifiable analytical methods for trade selection.

Recently, since around the first of June, the news from many "chart based" prognosticators is that the "markets have turned down". We commented on this news in our July 5 Newsletter (link on CISCO main page). Since then the USA has been involved in the 'Debt Limit' controversy which presumably negatively affects markets. The question is: "are the economists, chartists and other pundits correct in inferring that bearish markets should dictate day-trader's behavior"? A related question is "can you reliably measure if they are right or not"?


cont'd.......
Markets are not efficient, rather they are effective - Jones
Cont’d……Predictions from Charts, A Critique


Our approach today is to use the scientifically valid methods of observation, experimentation and verification. We want any conclusions reached to be supported by actual data. The path we will follow is based on Auction Market Value Theory. The primary principle (from observation) is that markets have two phases: they "pause" and a pause is terminated by a "run" (breakout). The run ends in a pause and the cycle continues. A pause is characterized by price clustering about a center (value). This cluster has an effective range, which the "market dynamic", determines. The 'value' concept is based in principle on the Market Profile value area, but practically, unlike the Profile, measuring value requires two or more days of data (found by experiment).


CISCO has developed a methodology(AMVT) for determining the condition of a market (pause or run) and a method for identifying the change when a market moves from a pause to a run (breakout) or from a run to a pause (congestion). In the following we will apply these tools to answer the question implied by the chartists: "If the charts show a directional bias for the market as a whole, is that information useful and important to a day trader?"

Cont’d……
Markets are not efficient, rather they are effective - Jones
Cont’d……Predictions from Charts, A Critique

Limitations
Analysis of complex systems like Auction Markets depends upon finding the "measureables". You could not measure the individual opinions of all traders in a market (even if you could, it would be always changing, and sometimes pretty fast). We can measure the clustering that defines a balanced market. This is done by the Overlay Demand Curve/Merged Profile. Value analysis depends on finding the market distribution. There must be adequate data for a representative measure: a market with just a few ticks per day tells little. The Meta-Profile and it's predecessor the Market Profile assumes the statistical distribution. There must be enough data (samples) to permit application of the formula.

The early profile work of Steidlmayer (ca 1985) took place under the four conditions:
1. The market was controlled by the 'Trade', the public was a small part.
2. It was all pit trading (with benefits to the members, none to the public).
3. Commercials, who could control the market, were physically visible in the pit.
4. Steidlmayer's Profile chart was an adjunct to the other information he picked up.


The Markets of Today
1. Automated trading cuts costs, improves fairness in fills.
2. Public dominates, a fairer marketplace. Public has up-to-date data.
3. Better data availability means much greater potential for 'data mining'.
3a. Computer based trading is possible for relatively small traders.
4. Extensive pre-market analyses is now possible for trading strategies.

Cont’d……
Markets are not efficient, rather they are effective - Jones
Cont’d……Predictions from Charts, A Critique

Finding Markets to Trade and How They Performed

Theory recognizes two primary market conditions: pause and run. Runs have many forms, but pause has only one major form, a cluster.

(Note: This Daily Tabulation (shown in the tables) is easily be done yourself, and is an everyday "practice" performed Daily for traders who actively trade using AMVT principles)

Table 1. below is an abbreviated EOD report of balances for next day's trading. Balance limits, UPPER & LOWER, give the range of the balancing cluster and an estimate of risk.

Table 2. A second, companion report, at the end of the next day's trading lists the markets that broke out and how far each market moved.

Cont’d……
Attached Image(s) (click to enlarge)
Click to Enlarge Name: table 1.JPG Size: 76 KB
Click to Enlarge Name: table 2.JPG Size: 85 KB
Markets are not efficient, rather they are effective - Jones
Cont’d……Predictions from Charts, A Critique

How Can You Tell if the Prognosticators are Right or Not?

The data in the above two tables is the sort that allows us to evaluate the implicit expectation that "The markets have turned down (and you should look for short side trades)". In our Newsletter of July 5 we concluded there was little difference between return on the longs and on the shorts over the June period (in which the general down direction was well developed).

A survey of active markets' breakouts directions following the procedure shows: see picture

Directional Results
In this professed period of down markets we see more upside breakouts than down. Adhering to the 'go with the market' dictum would have kept you out of 2/3 of the trades. On average, there is little difference between longs and shorts in the potential of a trade, as we will see.

Were the market pundits wrong on the market's direction? Apparently not, UU (emini SP) high on May 2 was 136450, low on August 2 is 124675, an 8.6 percent drop.

Were the pundits wrong in assuming the 'go with the market' dictum? We think not. Their problem is in defining what market they mean. They are not examining the market from a day trader's perspective.


Cont’d……
Attached Image (click to enlarge)
Click to Enlarge Name: table 3.JPG Size: 27 KB
Markets are not efficient, rather they are effective - Jones
Cont’d……Predictions from Charts, A Critique

Macro and Micro TimeFrames
So what is the problem? Can we be comparing apples to oranges? The general trend of a market as read from a chart normally gives a longer term picture.The chart covers several months and each day, even if the chart is that detailed, tends to meld into the general drift. This sort of display sacrifices the short term detail, exactly the time frame of interest to the day trader.

The typical chart may be considered a Macro time display. The longer timeframe trader (swing trader) may be satisfied with such a market picture. But even that type trader has to get in and out and that is a short time-frame question.

Most traders operate on inter-day time-frames, some within seconds but most no longer than the current day. This can be thought of as a Micro timeframe. The above end-of-day balances give the day-trader a starting point for market analysis on a known quantity--a balance with a well marked breakout price. Also included on this display is a risk estimate, 12.5 percent (1/8) of the balance price range.

Are the Chartists Correct in Their Directional Predictions?

The evidence for Macro (chart) directionality being a valid predictor of market directionality for day traders is negative. The consequence of this finding is far reaching. Probably well more than half the active day traders rely to some degree on chart information. Any time a trade is chosen on the basis of chart directionality, that choice is equivalent to flipping a coin. There is no 'edge' to a coin flip. So 'chart direction' as a guide to trading is not advised! In other words, the Macro market direction offers no help to the day trader.

Cont’d……
Markets are not efficient, rather they are effective - Jones
Cont’d……Predictions from Charts, A Critique


The U.S. Debt Limit Market Drop July 21 - August 2 2011
Over the eight days of final negotiations over the changing of the US debt limit, the Dow dropped steadily from 12,784.41 to 11,866.62. This is a real life laboratory for testing the thesis that day trading opportunities follow the general market trajectory.

In Table 4 below we have posted the day-by-day next day break-out market consequences, directions of the breakouts. The long/short count shows 42 longs and 42 shorts. In a decidedly down Macro market, the day trade consequences of the down Macro market is not apparent.


Cont’d……
Attached Image (click to enlarge)
Click to Enlarge Name: table 4.JPG Size: 31 KB
Markets are not efficient, rather they are effective - Jones
Cont’d……Predictions from Charts, A Critique

Serendipity, Being in the Right Place at the Right Time
This report was just being completed when on, August 4 the emini-SP fell out of bed.

Table 5. Report Aug 03, 11: UU and NQ Balances Before the Collapse

Two entries of interest in Table 5 are the indexes, NQ (Nasdaq) and UU (emini SP). A trade short for the 5 day Overlay for NQ, a breakout, would have been triggered at 07:46 at 227300 or below, ending at 15:15 at 221250 (+ 6050). UU would have been triggered short around 07:15. At the close the trade was up about 3825. No one (that I knew of) was predicting such a remarkable move.

Were we just lucky?
No. Serendipity works if you are not asleep at the switch.


Full article includes raw data & research details http://www.cisco-futures.com/chat_ioamt_110825.html
Attached Image (click to enlarge)
Click to Enlarge Name: table 5.JPG Size: 223 KB
Markets are not efficient, rather they are effective - Jones
Hello,
Since last week iam feeling it bit difficult to get the feel of what the markets are trying to do.
As you can see, daily profiles making higher highs and higher lows (VA higher, POC higher) most of the times its due to Gap Up open (RTH sessions only) and as there was some inventory imbalance (Intraday)which was corrected by this sudden Gap Up/Down, since last week its giving a false breakout and widening the 5Day ODC on both sides.

Attached Image (click to enlarge)
Click to Enlarge Name: NIFTYF Begining.png Size: 246 KB


I was eager to see the BO happen and quietly prices made some moves which negated the Breakout. But since last two days it has confirmed the BO and moved consistently but quietly, Vty is average(Not indicating demand responding to higher prices), TFF close to worse (Yday: 4.15 were as 90 day average is around 2.5), only confirmations is from the VA and POC shifting up.

Attached Image (click to enlarge)
Click to Enlarge Name: NIFTYF.png Size: 45 KB


Iam trying to get more info from the data by using Flow measurement, here iam failing to understand the calculation for F% (Att. Direction). I feel the attempted direction is more graphical visualization than mathematics. But I could get a mathematical idea of how successful the market was in the course of its attempted direction.

I searched through CISCO website and I could find this:
Attached Image (click to enlarge)
Click to Enlarge Name: 111.PNG Size: 6 KB


What Iam trying to do is: F% = ((Close-POC)/High-Low))*100

Also there is lot of mention about decoding commercial and other participants activity by understanding volumes in relation with Value area. Presently that part of information is not accessible to me due to platform constraints.

If you could help me on this,like by posting the Flow measurement Excel template and methods of understanding different participants activity you use in your analysis & calculation methods for the same, it would be kind of you.

Regards

Santosh.
If it was easy, everyone would do it.
  • Post #190
  • Quote
  • Edited 12:07pm Aug 14, 2014 11:56am | Edited 12:07pm
Quoting santoshdts
Hello, Since last week iam feeling it bit difficult to get the feel of what the markets are trying to do. As you can see, daily profiles making higher highs and higher lows (VA higher, POC higher) most of the times its due to Gap Up open (RTH sessions only) and as there was some inventory imbalance (Intraday)which was corrected by this sudden Gap Up/Down, since last week its giving a false breakout and widening the 5Day ODC on both sides. {image} I was eager to see the BO happen and quietly prices made some moves which negated the Breakout.
First, have you read Mind over Markets? or VBPT? From the looks of your charts I get the impression you're mixing the two concepts. Cisco recommends reading MOM as a prerequisite to apply AMVT concepts.

MP daily profiles is a day local concept. Steidlemayer was a floor trader "local" participant. Pit traders are short tf traders. They they create volume activity & liqudity in the intraday tf. However, they dont hold positions overnight. Only positions that are held overnight are create true demand and determine the direction of price.
Pete & Daltons perspectives come from a day TF perspective. They start flat and end the day flat. So many of the MP concepts & methodologies and trades taken were/are to be exited at the close. There is no concept of holding overnight.

There is much confusion when people use try to apply these short tf intraday MP concepts to the longer TF Overlays. Market profile methodology & concept arent designed for multiday positions. MP concepts work best for traders who start & end the day flat.

I also see you mention a BO. Depending on your perspective. AMVT overlays recognizes 2 conditions "Balance or trend". Break out from where? Im guessing from your chart you mean a BO from what? Value?
This is where I think there is confusion. The 70% VA concept from MP (only applies to the daily profile) Overlays determining a balanced market using Tpo count to determine balance. Not the 70% value area of the overlay. From an overlay perspective, the market has already been trending, there is not a Balance to break out of. Higher TFF suggests the auctioning/ rotating. To me it looks like trend is slowing or coming to an end.

Overlays dont have or use a 70% value area, a balance area is a single distribution of 3 tpos. Maybe I'm not understanding your display. But to me it looks likes a market that has been trending and TFF indicating that price is begining to Slow. Maybe Im not correctly understanding your charts.
Attached Image(s) (click to enlarge)
Click to Enlarge Name: va vs. brackets.jpg Size: 195 KB
Click to Enlarge Name: balance.jpg Size: 105 KB
Markets are not efficient, rather they are effective - Jones
I think Ive confused you by posting many charts with related observations to the charts mixed up..
Apologies!

Ill try to put it one at a time since this began.

Thanks for the reply.

Santosh
If it was easy, everyone would do it.
Quoting santoshdts
Iam trying to get more info from the data by using Flow measurement, here iam failing to understand the calculation for F% (Att. Direction). I feel the attempted direction is more graphical visualization than mathematics. But I could get a mathematical idea of how successful the market was in the course of its attempted direction. I searched through CISCO website and I could find this: {image} What Iam trying to do is: F% = ((Close-POC)/High-Low))*100 Also there is lot of mention about decoding commercial and other participants activity by understanding...
Its easier to ask is, the Close higher or lower than the POC by the amount of 20% of the days range......

I think you did it correctly I use

F%=(Close-POC)/(High-Low)*100
=IF(POC>=(High-Low*20%)>Open ="Up",
=IF(POC<=(High-Low*20%)<Open ="Down"
Anything else = "Stable"

Att dir uses 2 references
POC and Rotation factor
also it carries more weight when both references agree.......
Markets are not efficient, rather they are effective - Jones
Thanks for the Replies.

I Think Thjis is where I am Messing up "There is much confusion when people use try to apply these short tf intraday MP concepts to the longer TF Overlays. Market profile methodology & concept arent designed for multiday positions. MP concepts work best for traders who start & end the day flat. " & TPO counts at the EOD. This is where i am unable to read the daily auctions(Minor Auction ?) first and then relative to the multiday Overlay.

As far as the Multiday balance and Breakout is concerned I once again try to put up my view:
Attached Image (click to enlarge)
Click to Enlarge Name: NIFTYF 24-31.png Size: 48 KB

My confusion and Misunderstanding began with the choppy move as shown in the chat above.
Attached Image (click to enlarge)
Click to Enlarge Name: NIFTYF 01-07 ODC.png Size: 51 KB


since the Gap down opening and close of the following day above the Node of the Gap day profile. The Index is trending Up.
I Think in this attempt i was able to put up my view correctly.


Regards,
If it was easy, everyone would do it.
Quoting santoshdts
Thanks for the Replies. I Think Thjis is where I am Messing up "There is much confusion when people use try to apply these short tf intraday MP concepts to the longer TF Overlays. Market profile methodology & concept arent designed for multiday positions. MP concepts work best for traders who start & end the day flat. " & TPO counts at the EOD. This is where i am unable to read the daily auctions(Minor Auction ?) first and then relative to the multiday Overlay. As far as the Multiday balance and Breakout is concerned I once again try to put up my...
Ok, I think I have an idea what the issue is, again trying to apply MP concepts to AMVT Overlays. MP methodology compares one day value relative to the previous days profile. Not the case for Overlays. Today the 5 day overlay is a snapshot/picture (profile) of the 5 day distribution. Tomarrows 5 day overlay will be a snapshot/of the current 5 day distribution. As each day passes one day is eliminated and the most recent day added. Everyday the overlay displays the last 5 days data. Everyday that snap shot changes.........

In analyzing the data you compare yesterdays 5 day overlay reference points, to today's 5 day overlay reference points. Its what you could call a "Rolling profile" Everyday you will be looking at a profile of the last 5 days data............
All this can be found in VBPT........
Markets are not efficient, rather they are effective - Jones
Quoting santoshdts
Thanks for the Replies. I Think Thjis is where I am Messing up "There is much confusion when people use try to apply these short tf intraday MP concepts to the longer TF Overlays. Market profile methodology & concept arent designed for multiday positions. MP concepts work best for traders who start & end the day flat. " & TPO counts at the EOD. This is where i am unable to read the daily auctions(Minor Auction ?) first and then relative to the multiday Overlay. As far as the Multiday balance and Breakout is concerned I once again try to put up my...
Your 5 day overlay is a 5 day distribution 5 days ago (old data)
Attached Image (click to enlarge)
Click to Enlarge Name: Untitled picture.png Size: 43 KB
Markets are not efficient, rather they are effective - Jones
Quoting mzvega
{quote} Your 5 day overlay is a 5 day distribution 5 days ago (old data) {image}
here's an example..........
Attached Image(s) (click to enlarge)
Click to Enlarge Name: 5 day of profiles.jpg Size: 350 KB
Click to Enlarge Name: 5 day.jpg Size: 417 KB
Markets are not efficient, rather they are effective - Jones
​
TPO Overlays vs. Volume Overlays
VBPT p 64-71

Overlay Demand Curves are normally generated from volume data in the LDB reports or from TPOs. In principle, volume offers the closest link to demand; volume is a direct count of the amount traded in the market place. As noted earlier, volume at price is currently released only by the Chicago Board of Trade and the Chicago Mercantile Exchange. Volume-based Overlays cannot currently be generated for any other exchanges.

Cleared volumes are not necessarily a direct measure of demand. Much of a day's trading is between floor traders (Cti1), whose primary function is to provide liquidity to the market. Most floor traders go home with no positions and thus have little effect on demand. Floor traders are typically responsible for about half the total volume reported. So a sizeable fraction of the volume could be giving either the wrong signal or no signal at all. Tick volume suffers from the same malady.

TPO counts tend to be somewhat more straightforward in measuring activity. At price extremes where the public tends to dominate trading, the public is filled by the floor trader.

As seen in Chapter 3, Figure 3-1, trading at 10621 is half floor, half public; the floor is servicing the public.

At the middle prices, where the floor is very active, limiting the TPOs to one per half-hour (at each price) tends to ensure that value is involved when a group of TPOs is created.

Comparing Overlays from both LDB volume data and TPOs for the same trading period brings the differences into focus. figures 4-1 and 4-2 compare TPOs with volume for T-bonds, with Figure 4-3 overlaying the TPOs onto the volume for added emphasis. A much more severe case is examined in Figure 4-4, where there are added problems from adjacent prices being differentially valued by traders.


At the middle prices, where the floor is very active, limiting the TPOs to one per half-hour (at each price) tends to ensure that value is involved when a group of TPOs is created.

Comparing Overlays from both LDB volume data and TPOs for the same trading period brings the differences into focus. figures 4-1 and 4-2 compare TPOs with volume for T-bonds, with Figure 4-3 overlaying the TPOs onto the volume for added emphasis. A much more severe case is examined in Figure 4-4, where there are added problems from adjacent prices being differentially valued by traders.
Attached Image(s) (click to enlarge)
Click to Enlarge Name: Capture.JPG Size: 266 KB
Click to Enlarge Name: 4-4.jpg Size: 113 KB
Markets are not efficient, rather they are effective - Jones
TPO and volume Overlays are different ways of measuring the same thing. The general structure is obviously the same, but with volume the structure is more “ragged” (compare, for example, the Overlays at prices 10603, 10527, 10515 in figure 4-1 and 4-2). The smoothness of the TPO Overlay is an even larger advantage in markets such as the grains. Wheat is shown in Figure 4-4 with a volume Overlay on the left, and its TPO counterpart on the right.

A basic difficulty caused by the asymmetries of volume Overlays is their effect on determining value. Different treatments give different answers. In Figure 4-3, the peak volume in T-bonds is at 10603, on the edge of the distribution, while peak TPO count came at 10531, much nearer the center of the principal lobe of the distribution. For wheat in Figure 4-4, the TPO Overlays have a smooth, well-defined distribution compared to the volume distribution, which could easily be interpreted as several distributions.

An observed characteristic of the volume Overlays is that the high-volume price often acts as a barrier within a day. Prices tend to reflect off the barrier. As long as market condition does not change, the high-value Point behaves as a resistance to prices below it and a support to those above.

TPO Overlays are automatically smoothed by time. Only one event can be measured at a Particular price in any half-hour period. Multiple TPOs at a price indicate that the market returned to that price region throughout the time frame of the Overlay. Consequently, TPO distributions are smoother and easier to read and use. Furthermore, all markets can be analyzed with TPO Overlays, whereas volumes are available only for the Chicago Board of Trade and the Chicago Mercantile Exchange. Another benefit of the TPO Overlay is that it can be updated during the day. TPOs for the current day can be added to an Overlay as they develop
Attached Image(s) (click to enlarge)
Click to Enlarge Name: 4-4.jpg Size: 113 KB
Click to Enlarge Name: Capture.JPG Size: 266 KB
Markets are not efficient, rather they are effective - Jones
NICE work Claudia. Very illuminating to learn more about AMVT from your point the view & reiterating the key points of it too.

I Have not been here for awhile didn't know you have started a thread in FF with this topics solely in mind. Great work.
Quoting Evak
NICE work Claudia. Very illuminating to learn more about AMVT from your point the view & reiterating the key points of it too. I Have not been here for awhile didn't know you have started a thread in FF with this topics solely in mind. Great work.
What a wonderful surprise it was to see your post this morning!!! I started the thread in hope that maybe I could lure you and possibly a few other senior members from the old thread. You're the only person I know who loves & understands this subject as much, if not more than I. I figured if I could keep the thread on topic, you'd show up eventually. Its great to see your post! I hope you continue to post, share your thoughts & insights as well. I only post here occasionally, as there is no interest in the actual "Analysis" of either MP or AMVT.

Well, anyway I hope you can find the time to post more often, you are a valuable contributor to any thread, no matter where you post...................I hope you decide not to be a stranger. With Bandung's new TPOChart2.1 I'm busy trying to expand my Excel sheet analysis into a database to build my own "bracket Screen" and analysis tools similar to the tools used at Cisco for Futures data, but it will be Forex in my case........

Anyway, don't be a stranger!!!!!!
Markets are not efficient, rather they are effective - Jones
0 traders viewing now