Price & Volume & Other Things (possibly indicators too at some stage...)
- Thread starter TheBramble
- Start date
- Watchers 8
TheBramble
Legendary member
- Messages
- 8,394
- Likes
- 1,170
Yet another attempt to gloss over what is instinctively known without any attempt to empirically define. Not that this is a hanging offence, but you're not addressing the thrust of what this thread is about. Not surprising from another dbp fan, but it was to be expected that this thread would (initially at any rate) attract that crowd.sulong said:
None of what you state above moves one iota closer to empirical definition - which is what this thread is about.
It's not about simple ideas being complicated - it's all about defining what it is we instinctively take as simple. Which is paradoxically quite hard.
It's like asking someone to explain electricity. They say, "Well, you flick this switch and the light comes on. Flick it the other way and it goes off. That's electricity".
I'm happy to have thumbnails, wordy definitions or movies (Des - where are you?), but the same old tosh - textbook definitions or regurgitated guru output really isn't going to cut it.
As for 'only visual elements'...Lord! Some people work a lot better with visual than digital (words). A tad myopic sulong...
barjon
Legendary member
- Messages
- 10,752
- Likes
- 1,863
Tony
Whilst it won't put us off since it's what this thread is about I'm not at all certain that trying to pin down empirical definitions of high specificity is either possible or advantageous.
Some customs officers regularly and consistently catch more smugglers than their colleagues. Even after comprehensive de-briefing they cannot say why in measurement terms even though many of the salient factors that added to their suspicion can be teased out (albeit even that is difficult). It's all put down to Revenue Nose which is the subconscious application of acquired knowledge which itself is a bit of a mystery because some of them are that good from the moment they put on the uniform.
I remember Bonsai (who rarely made a wrong call) often used to call a close to his trade because, in his famous phrase, it's just sitting there looking at me. I suspect he would have been hard pressed to give you an empirical, specific and measured definition.
We'll have a go though
good trading
jon
Whilst it won't put us off since it's what this thread is about I'm not at all certain that trying to pin down empirical definitions of high specificity is either possible or advantageous.
Some customs officers regularly and consistently catch more smugglers than their colleagues. Even after comprehensive de-briefing they cannot say why in measurement terms even though many of the salient factors that added to their suspicion can be teased out (albeit even that is difficult). It's all put down to Revenue Nose which is the subconscious application of acquired knowledge which itself is a bit of a mystery because some of them are that good from the moment they put on the uniform.
I remember Bonsai (who rarely made a wrong call) often used to call a close to his trade because, in his famous phrase, it's just sitting there looking at me. I suspect he would have been hard pressed to give you an empirical, specific and measured definition.
We'll have a go though
good trading
jon
TheBramble
Legendary member
- Messages
- 8,394
- Likes
- 1,170
barjon said:
Everything is possible and as for advantageous, well, the exercise may not yield a definitive result, but there will inevitably be benefits in expending the effort (personal opinion).
The Bonsai quote is a perfect example of what it is I'm trying to crack. You (we/one) has to go through the process of defining what it is we do to make successful trades. We (it would seem) do not formally acknowledge them, record them or can even easily verbalise them - but we must know what they are or we wouldn't be making the trades.
I'm having difficulty myself with digging into what it is I do, I thought my evolving catharsis (oooh!!!!) may not only help others, but also myself in pinning it down.
JumpOff
1
- Messages
- 702
- Likes
- 14
TheBramble said:
Really Bramble! You are swiping with a mighty broad brush...... JumpOff wipes the spattered tar off her ears.... Sulong may have reasons for thinking you are examining your own (trader's) navel that have nothing to do with the fact that he is what you call "a dbp fan."
And I agree with you that he rather neatly sidestepped the essence of your thread. - So lead on, oh thorniest one......! I suspect that this thread will lead to useful discussion.
JO
Tony,
What about including the difference between today's high and yesterday's high (as today's can either be the same, higher, or lower than yesterdays high) and also maintain average of the differences? This could give an idea of potential penetrations of previous day's high or lack of penetration when they occur. Of course, breakouts would nullify this but most of the time we don't have break out senarios.
PT
What about including the difference between today's high and yesterday's high (as today's can either be the same, higher, or lower than yesterdays high) and also maintain average of the differences? This could give an idea of potential penetrations of previous day's high or lack of penetration when they occur. Of course, breakouts would nullify this but most of the time we don't have break out senarios.
PT
Last edited:
If one is keeping a running tab of the penetrations of the previous days high then that average penetration is what it tends to penetrate except when there are breakouts. Since most of the time we don't have breakouts (and this can be confirmed by looking at any daily chart) we can expect average penetrations of the previous day's high when there are in fact penentrations of the previous days high. Breakouts would invalidate the averages because they are not the average/normal price action. Does this explain what I mean?JumpOff said:
PT
JOJumpOff said:
It may or may not return into the days range. For instance, it may open on the low and trade up closing on a high that penetrated yesterdays high at the end of the day. Or, it may open high trade higher, penetrate yesterday's high then retrace trading down and finally close near the low of the day after having made the high penetration. Strong breakouts that carry price up far are usually obvious...made on strong volume ...with weaker volume on retracements before resuming the breakout again. However, what about a penetration of yesterdays high that was not done on a true breakout with strong volume? In such a case it does help to have a running tab of the tendency of a stock or futures to penetrate the previous days high when it is not a breakout. Why? This gives a potential high for the day and a possible shorting point for the day. Or a potential point for selling longs previousley aquired. The point is it helps to know where it "may" stop at when it is not a true breakout. Naturally, the final determining factor has to be be what the tape is saying at the moment but I find it does help to have a "view" on any potential high.
PT
TheBramble
Legendary member
- Messages
- 8,394
- Likes
- 1,170
Good stuff. Just what I was hoping for - innovative thought in the relative comparisons of basic price/volume data.pttrader said:
OK. So what could we term (however temporarily) such a measure of Highs and Lows?
And what might their significance be?
I'm not too keen to start thinking about averages as useful as they are - they're one degree removed from the basic price/volume data and I suspect we'll have enough on our plate with just the basic panoply as we get deeper into these analyses.
TheBramble
Legendary member
- Messages
- 8,394
- Likes
- 1,170
You're talking about the Price 'movement' as a function of time. What you introduced this issue with was the high-EST (and I guess by extension the low-EST) points the price reaches in each bar in relation to the high-est/low-est in previous bars. What does that static measure mean?pttrader said:
If you divide todays high by yesterdays high the result can be defined as "volatility" on the upper end of the range. However, I would prefer for simplicity's sake to just keep the averages of such a measure in price terminology because that is how we buy and sell. We don't buy and sell in terms of percentage difference in volatility.TheBramble said:
What does it mean? For me it simply means the tendency for a penetration (or lack of a penetration) to reach a certain price level. It could be expressed mathematically as volatility in positive or negative terms on the upper end of the daily range but again it seems to me to be best to keep it in price terminology.
It could also be expressed as momentum on the upper end of the daily range. That is; today's high minus yesterday's high = momentum. Although one day momentum is usually expressed as today's close - yesterday's close. But one could measure the momentum just on the upper end. Rate of change and momentum in all practical terms are the same measurment but made with different calculations. The end result is the same for practical purpose.
However, again so as to not get too far off track I still prefer to simply express it as a tendency in price change in terms of price itself on the upper end. By extension on the lower end also.
PT
Bram,
The 3 Functions of analysis, may be described under three headings, Descriptive, Selective, and Critical.
Descriptive analysis, consists of marshalling the important facts relating to an issue and presenting them in a coherent, readily intelligble manner. This will vary I suppose from individual to individual. A more penetrating type of description seeks to reveal the strong and weak points in the position of an issue.
Selective analysis, goes further and expresses specific judgements of its own. It seeks to determine whether a specific issue should be bought, sold, retained, or exchanged for another.
Critical analysis, is the evaluation of your analytical judgements. The analyst is concerned, therefore, with the soundness and practibility of the standards of selection. This would require an objective assessment of the results and how obtained.
There are of course three obstacles to success. Incorrect or faulty data, probably not an issue of any great import. Uncertainities of the future, and Irrational behaviour of the market.
The element of future change, may subvert the conclusion drawn by a warranted analysis of the facts, and by the apparent prospects may be vitiated by new developments.
This raises the question of how far is it the function of analysis to anticipate changed conditions. It is manifest however that future changes are largely unpredictable, and that analysis must ordinarily proceed on the assumption that the past record affords at least a rough guide to the future.
The third handicap is found in the market itself. In a sense the market and the future present the same type of difficulties. Neither can be predicted or controlled by the analyst, yet his success is largely dependant upon them both.
It is this double jeopardy that brings into question.........what exactly is the analyst analysing?
Is it the security, or is it the interest in the security?
Analysis of the incorrect factor, and or the belief one is analysing one, while in reality analysing the other, will lead to confusion, and unimpressive results.
As your thread has demonstrated yet again, is that many traders fundamentally do not understand what they are actually attempting. This is bourne out by the total lack of definition as to their structure, philosophy, and execution of a plan.
The minutae of trading will only help those who have established an understanding of the gross. Microscopic focus on the detail, without the latter, is joining an Olympic sprint, after one afternoon jog in the park.
cheers d998
The 3 Functions of analysis, may be described under three headings, Descriptive, Selective, and Critical.
Descriptive analysis, consists of marshalling the important facts relating to an issue and presenting them in a coherent, readily intelligble manner. This will vary I suppose from individual to individual. A more penetrating type of description seeks to reveal the strong and weak points in the position of an issue.
Selective analysis, goes further and expresses specific judgements of its own. It seeks to determine whether a specific issue should be bought, sold, retained, or exchanged for another.
Critical analysis, is the evaluation of your analytical judgements. The analyst is concerned, therefore, with the soundness and practibility of the standards of selection. This would require an objective assessment of the results and how obtained.
There are of course three obstacles to success. Incorrect or faulty data, probably not an issue of any great import. Uncertainities of the future, and Irrational behaviour of the market.
The element of future change, may subvert the conclusion drawn by a warranted analysis of the facts, and by the apparent prospects may be vitiated by new developments.
This raises the question of how far is it the function of analysis to anticipate changed conditions. It is manifest however that future changes are largely unpredictable, and that analysis must ordinarily proceed on the assumption that the past record affords at least a rough guide to the future.
The third handicap is found in the market itself. In a sense the market and the future present the same type of difficulties. Neither can be predicted or controlled by the analyst, yet his success is largely dependant upon them both.
It is this double jeopardy that brings into question.........what exactly is the analyst analysing?
Is it the security, or is it the interest in the security?
Analysis of the incorrect factor, and or the belief one is analysing one, while in reality analysing the other, will lead to confusion, and unimpressive results.
As your thread has demonstrated yet again, is that many traders fundamentally do not understand what they are actually attempting. This is bourne out by the total lack of definition as to their structure, philosophy, and execution of a plan.
The minutae of trading will only help those who have established an understanding of the gross. Microscopic focus on the detail, without the latter, is joining an Olympic sprint, after one afternoon jog in the park.
cheers d998
Last edited:
We could call it HH (for high,high) or HE (for high end) as a measurement of that unit on the upper end.TheBramble said:
LL for lower end (low,low) or LE (low end) as a measurement for that particular unit on the lower end of the daily range.
I might say not using an average of the figures and not extending that average out onto a forecast for tomorrow we will find that the forecast will be too unstable and not accurate enough to be of much practical use. On the other hand, to add the average figure of the last 10 days to the high of today to get an extension with a view on tomorrow's high (providing it does in fact penetrate today's high) is more stable and tends to be more accurate than just adding yesterday's figure to todays high to get a forecast for tomorrows high. Why? Because yesterday's high could have been a spike or even a break to the upside something that would throw the figures too far out of wack to be of practical use. For me averages are the way to go. In addition, we are still talking of averages of the "tape" so we are still dealing with actual price.
One other point I might make is that for this extension forecast to be looked at in terms of the "strenght of the penetration" it must be compared to the volume when the penetration was made. This is simply "tape reading". Price tells what happened. Averages tell what may happen. When "whatever" does in fact happen, then volume tells "how" it happened.
PT
Last edited:
timsk
Legendary member
- Messages
- 8,596
- Likes
- 3,320
Hi pttrader,
On behalf of all the dumbos who are trying to follow your posts but can't quite keep up with the pace, any chance of posting an annotated chart illustrating the points you're making? I'd really appreciate it and I'm sure others would too.
Cheers,
Tim.
On behalf of all the dumbos who are trying to follow your posts but can't quite keep up with the pace, any chance of posting an annotated chart illustrating the points you're making? I'd really appreciate it and I'm sure others would too.
Cheers,
Tim.
TheBramble
Legendary member
- Messages
- 8,394
- Likes
- 1,170
At whom this thread is unashamedly pitched....ducati998 said:
I have been stressing all the way through that what I'm hoping is for other successful traders to examine what it is they do in relation to these (and other yet-to-be-discovered?) factors/qualities.
I'm also absolutely happy for those still experimenting/learning to contribute on a hypothesis basis.
After all, you never know what you don't know....
I think pttrader's idea has potential in illustrating the strength of a trend and predicting where the day's high might be but I would have thought that its results would overlap with a number of existing indicators/oscillators. It is a good idea though because it gives a simple quantitive approach to the basic peak and troughs method of technical analysis. I agree that you have to use averages to make this idea work.
As I understand it, the idea is:
HH = {(h2-h1) + (h3-h2) + (h4-h3) + ........ + (hn - hn-1)}/n
where hn is the highest price on day n.
Alternatively, instead of subtracting the previous period's high, you could try subtracting the high from all periods to give a slightly different statistic.
You can add HH to yesterday’s high to give a prediction of where today’s high might be.
To go into more detail, I think it has two main applications:
1) Gauging the strength of a prevailing trend.
One should not limit things to daily periods - any time-frame should be fine to test. However as we all know, even in a strong uptrend, you never get consistently higher highs. Therefore, for the purpose of gauging momentum I think the statistic has to assign lower highs (in an uptrend) a negative value and incorporate these into the average. If we monitor an n period moving average of this statistic, I suspect that we would find that when it is positive and increasing, the market has strong enough momentum that more often than not will prove to be a good buying point. As it starts to decline we would get a signal to scale out of long positions.
Obviously a lot of backtesting is required for us to derive a trading rule, but it is likely to give us some decent tradable rules.
An extension of this would be to include the movement of the day's low into the statistic (even when investigating an uptrend). As I said before, however, it is not clear to me whether this statistic will yield significantly different results from some existing indicators. I am not that informed about many of these - maybe someone else knows of such an indicator.
2) Forecasting where the day's high will be.
If the last 4 trading sessions saw the previous high surpassed by 10 ticks, then it would be reasonable to assume that this session's high might also be 10 ticks higher than last session's high. Therefore, you could try taking a short from that level. (Or of course, go long with that target in mind). Obviously the differences between highs will not be that consistent however. Therefore, we also need to bear in mind the variability of these differences. When this second statistic is relatively low, then the target buying/selling strategy might be attractive. You could use the variability statistic to help guide you with a stop level, although we obviously would need to do further investigations to come up with a sensible target for any retracement.
I see some potential in both of these applications for various liquid instruments. e.g. EURUSD, Bund, equity indices. However, it only captures a small amount of information from the previous trading pattern that alone is unlikely to have that much predictive strength. It would therefore probably have to be combined with some other indicators to develop a successful strategy. Unfortunately I do not have access to any decent data at the moment but I would be interested to hear if anyone has any success spotting any patterns.
If we get somewhere with this approach, a possible extension would be to keep a statistic for the amount of volume traded in the area above the previous day's high.
As I understand it, the idea is:
HH = {(h2-h1) + (h3-h2) + (h4-h3) + ........ + (hn - hn-1)}/n
where hn is the highest price on day n.
Alternatively, instead of subtracting the previous period's high, you could try subtracting the high from all periods to give a slightly different statistic.
You can add HH to yesterday’s high to give a prediction of where today’s high might be.
To go into more detail, I think it has two main applications:
1) Gauging the strength of a prevailing trend.
One should not limit things to daily periods - any time-frame should be fine to test. However as we all know, even in a strong uptrend, you never get consistently higher highs. Therefore, for the purpose of gauging momentum I think the statistic has to assign lower highs (in an uptrend) a negative value and incorporate these into the average. If we monitor an n period moving average of this statistic, I suspect that we would find that when it is positive and increasing, the market has strong enough momentum that more often than not will prove to be a good buying point. As it starts to decline we would get a signal to scale out of long positions.
Obviously a lot of backtesting is required for us to derive a trading rule, but it is likely to give us some decent tradable rules.
An extension of this would be to include the movement of the day's low into the statistic (even when investigating an uptrend). As I said before, however, it is not clear to me whether this statistic will yield significantly different results from some existing indicators. I am not that informed about many of these - maybe someone else knows of such an indicator.
2) Forecasting where the day's high will be.
If the last 4 trading sessions saw the previous high surpassed by 10 ticks, then it would be reasonable to assume that this session's high might also be 10 ticks higher than last session's high. Therefore, you could try taking a short from that level. (Or of course, go long with that target in mind). Obviously the differences between highs will not be that consistent however. Therefore, we also need to bear in mind the variability of these differences. When this second statistic is relatively low, then the target buying/selling strategy might be attractive. You could use the variability statistic to help guide you with a stop level, although we obviously would need to do further investigations to come up with a sensible target for any retracement.
I see some potential in both of these applications for various liquid instruments. e.g. EURUSD, Bund, equity indices. However, it only captures a small amount of information from the previous trading pattern that alone is unlikely to have that much predictive strength. It would therefore probably have to be combined with some other indicators to develop a successful strategy. Unfortunately I do not have access to any decent data at the moment but I would be interested to hear if anyone has any success spotting any patterns.
If we get somewhere with this approach, a possible extension would be to keep a statistic for the amount of volume traded in the area above the previous day's high.
Many stocks, futures, indexes tend to trend 3 to 7 days before a pullback. Many times it is simply a 3 or 4 day cycle. IMO this fact makes working this on a daily basis more feasible. However, if one decided to use it on an intraday basis I think it necessary to correlate it with the time of day. We usually get a first move early in the session. A pullback before noon. Sometimes another push in the prevailing trend. A dead spot. And then sometimes an afternoon another move that mirrors the morning move. Anyway, to use it intraday I do think one would need to keep in mind key times during the day when moves tend to be made. We get that dead spot because senior MM and specialists are gone to lunch and don't won't the jrs. letting things get too far out hand. Action picks back up around 1:30 p.m.- 2:00 p.m. eastern after lunch is well over with. Of course, I am dealing here primarily with U.S. stocks and indexes. Not sure how this would work in say Forex? With other countries stocks/indexes I suppose floor brokers...etc they have to eat too! It is right funny out we humans are so habitual and that it even reflects in the markets! Imagine eating affecting stock prices!
Negative values must also be included in the calculations.
PT
Last edited:
