Delta Volume - Cumulative Delta Volume | Page 6 | Forex Factory

Options

Go to
Permalinks
Print Thread

Similar Threads

Cumulative Delta Price Volume 618 replies

Need help to make Cumulative Volume Index indicator for MT4 70 replies

Interactive Brokers Cumulative Delta indicator suggestion 1 reply

Platform with Cumulative Delta and Real Time Volume Data 2 replies

Convert delta(EMA) -> delta(Close) 0 replies

Attachments: Delta Volume - Cumulative Delta Volume
Exit Attachments

Delta Volume - Cumulative Delta Volume

  • #101
  • Apr 7, 2016 1:10am Apr 7, 2016 1:10am
Quoting truelifeajf
sure, but it's predominantly ask "agitation". Let's not get caught up in the detail here. The point is, when is "agitation" buying / selling volume, and when is it spread adjustment? FWIW, there's 20% more Ask changes in that chart than there are Bid changes.
It's true that I do not have sufficient information about the broker's procedure for spread adjustment. It would be interesting to get such metrics.
  • #102
  • Apr 7, 2016 1:15am Apr 7, 2016 1:15am
I've noticed that there's always more ask movement than bid meaning always more "buy" volume than "sell".

I'm curious if any other users have noticed this.

If you ignore when the bid is unchanged, and only the ask moves (as people have discussed earlier), that's not a good solution either. If you need both bid and ask to move (up or down) to record a tick, then you may as well just record the movement of the bid price meaning you no longer have a bid / ask tick recorder.
  • #103
  • Apr 7, 2016 9:32am Apr 7, 2016 9:32am
Quoting Mingary
{quote} How does one use this wonderful toy ?
http://bit.ly/1NaO84J
practising coitus interruptus
  • #104
  • Apr 7, 2016 10:52pm Apr 7, 2016 10:52pm
Here's a tick chart for a 6 hour period on EURUSD starting today.

The differences in bid / ask are really quite varied between brokers.

Global Prime had 27,700 ticks whereas FXCM had 12,240 ticks. That's a big difference.

The two charts don't share the same timescale so they don't line up with each other, so you have to take that into account (they can't line up because in any given minute each broker will have a different number of ticks).

But you can see big enough differences.

If we attribute buy volume to the ask going up, and sell volume to the bid going down then attached are the 1 minute charts of the cumulative delta volume for each broker. Again, very different between brokers.
Attached Image(s) (click to enlarge)
Click to Enlarge Name: fxcm_ticks.png Size: 49 KB
Click to Enlarge Name: gp_ticks.png Size: 50 KB
Click to Enlarge Name: fxcm_minute.png Size: 35 KB
Click to Enlarge Name: gp_min.png Size: 34 KB
  • #105
  • Edited 2:25am Apr 8, 2016 12:21am | Edited 2:25am
A sample of LMAX volume Vs OANDA volume same time period, 1 M Euro

Attached Image (click to enlarge)
Click to Enlarge Name: VolOANDALMAXEUR.gif Size: 35 KB


WRONG ==> (The differences are obvious. It's difficult to justify using volume in the forex)
TRUE ==> ... and yet the price moves the same.
If the constant is the price, but the volume is the variable, then the reaction to price as reflected in the volume would be worth studying - if we assume that the volume is the footprint of what the traders are doing .. and your fx broker does not play games with the price feed.

On second look I did not take the time zone into consideration ==> now it's a better match. !

Attached Image (click to enlarge)
Click to Enlarge Name: VolOANDALMAXEUR2.gif Size: 34 KB
  • #106
  • Apr 10, 2016 4:14pm Apr 10, 2016 4:14pm
Quoting Mingary
{quote} It all depends on the frequency of the brokers' spread adjustment. Does it happen on every tick ? Also when the broker widens the spread, it applies for Ask and Bid. Brokers' adjustment: Ask goes up, Bid goes down = wide spread
I'm not sure I agree. We know spreads can widen hugely before news, Mondays, etc.

But if you overlay two charts from two different brokers, the charts are 99.9% identical.

The differences in spread are all (mostly?) in the Ask price going up and down.

If the bid price varied as much as the ask price then everyone's charts would look very different from each other.

There's always more "buying volume" (ask going up) than there is "selling volume" (bid going down).
The price will go up, and down, not necessarily in that order.
  • #107
  • Apr 11, 2016 3:19am Apr 11, 2016 3:19am
Quoting ForexFart
{quote} I'm not sure I agree. We know spreads can widen hugely before news, Mondays, etc. But if you overlay two charts from two different brokers, the charts are 99.9% identical. The differences in spread are all (mostly?) in the Ask price going up and down. If the bid price varied as much as the ask price then everyone's charts would look very different from each other. There's always more "buying volume" (ask going up) than there is "selling volume" (bid going down).
In MT4 the bid is the actual price, the spread is not wrapped around the market price. Therefore brokers tend to not mess with the bid, but they sure make a mess of the Ask, every chance they see they can make some bucks ..;
The bid is the only "true" price information in MT4
  • #108
  • Apr 14, 2016 7:51am Apr 14, 2016 7:51am
I have been away for a while, but now that I am back I must say that I am impressed with the critical discussion here lately. In trading, we should not take things for granted and we should try to get into the nuts and bolts of what is presented to us.
  • #109
  • Apr 14, 2016 8:01am Apr 14, 2016 8:01am
Quoting truelifeajf
If "volume" in forex is just the "number of price changes" in a given period, then the volume for each tick should be 1. And for 99% of cases, this is true. But for about 1% of ticks, the volume is 2. And for a really small percentage of ticks, the volume might be 13 or 60 or 88. Does anyone have an opinion about why we get such large volume on a mere price change, when volume is meant to be a single price change?
You are correct, Many ticks that come through on MT4 do not have any volume data associated with them, and many of the ones that do have a volume of 1. One what? one million, billion, trillion? Who knows. The fact that volume comes in as an integer may be a clue that perhaps it is a cumulative volume itself. So volume will only come through once a certain threshold is breached, and then the counter is set to "zero" again. This would mean that the volume coming through would be a combination of buy and sell volumes over the period from when the last volume data came through. I'm not 100% convinced that this is the case, but it may very well be.
  • #110
  • Apr 14, 2016 8:04am Apr 14, 2016 8:04am
welcome back. My biggest problem with saying ask up = buy and bid down = sell is there's always more ask movement that bid (and therefore more buying volume than selling volume) because of spread adjustments (which in my opinion affect the ask, because if the bid varied too much, charts between brokers would be different... and they're not)
The price will go up, and down, not necessarily in that order.
  • #111
  • Apr 14, 2016 8:10am Apr 14, 2016 8:10am
Quoting truelifeajf
Here's a tick chart for a 6 hour period on EURUSD starting today. The differences in bid / ask are really quite varied between brokers. Global Prime had 27,700 ticks whereas FXCM had 12,240 ticks. That's a big difference. The two charts don't share the same timescale so they don't line up with each other, so you have to take that into account (they can't line up because in any given minute each broker will have a different number of ticks). But you can see big enough differences. If we attribute buy volume to the ask going up, and sell volume to...
Yes, brokers have very different feeds. Oanda, for example, filters their ticks substantially. For every 1 tick on Oanda, you might get 3-5 ticks on another broker. Hence, any "nuts and bolts" view of volume must be broker-specific. Each broker is manages their books and risk independently, and they set their spreads accordingly. But, they have a massive amount of market information that we don't as retail traders, and it might be possible to study their pricing behaviour to gain insight into market conditions.
  • #112
  • Apr 14, 2016 8:12am Apr 14, 2016 8:12am
Oh, I absolutely agree it's worth looking into the data. It's just a shame that the delineation between spread adjustment and ask (buy) volume is completely unknown.
The price will go up, and down, not necessarily in that order.
  • #113
  • Apr 14, 2016 8:22am Apr 14, 2016 8:22am
Quoting ForexFart
welcome back. My biggest problem with saying ask up = buy and bid down = sell is there's always more ask movement that bid (and therefore more buying volume than selling volume) because of spread adjustments (which in my opinion affect the ask, because if the bid varied too much, charts between brokers would be different... and they're not)
Again, this is probably broker-specific. I have not noticed a difference in ask/bid ticks with my broker. And spreads widen both ways, which is especially noticeable at NY Close when I can get a weird "floating" bar.

At the end of the day, Its good that you are critical of this. It shows that you are thinking about the mechanics of the market, something that a lot of retail traders don't put a lot of thought into.
  • #114
  • Apr 14, 2016 8:56am Apr 14, 2016 8:56am
Attached are the two charts presented earlier (comparing bid and ask between two brokers), but scaled to the same number of ticks and overlaid on each other.

Actually the bid does vary as well which perhaps doesn't mean spread adjustment is ONLY the changing of the ask price. I guess the bid differences are so small that you don't notice the differences when comparing charts from two brokers.

Also, I guess so long as at the start and end of each minute the bid is the same between two brokers, what happens during the minute in terms of the bid moving due to spread adjustment doesn't matter. The chart can only "zoom" to the 1 minute timeframe and so differences in bid WITHIN the 1 minute would not be seen by traders... unless you look at a tick chart as per the attached.
Attached Image (click to enlarge)
Click to Enlarge Name: 2 broker compare.png Size: 98 KB
The price will go up, and down, not necessarily in that order.
  • #115
  • Apr 14, 2016 9:34am Apr 14, 2016 9:34am
Quoting ForexFart
Attached are the two charts presented earlier (comparing bid and ask between two brokers), but scaled to the same number of ticks and overlaid on each other. Actually the bid does vary as well which perhaps doesn't mean spread adjustment is ONLY the changing of the ask price. I guess the bid differences are so small that you don't notice the differences when comparing charts from two brokers. Also, I guess so long as at the start and end of each minute the bid is the same between two brokers, what happens during the minute in terms of the bid moving...
Its actually really cool to compare brokers in this way, and I'm fairly surprised that they matched up so well after adjusting for the number of ticks, meaning that overall both brokers had about the same number of ticks over the period even though their pricing differed. Its a bit difficult to tell with the resolution that I have here what is going on when the ask "flutters". It looks like both brokers are doing it at the same time. There is one spot where the spread widens due to the bid going down, and other spots where the spread widens due to the ask going up. There are even some spots where the bid from one broker is higher than the ask from the other! Are these ECN brokers?
  • #116
  • Apr 14, 2016 9:38am Apr 14, 2016 9:38am
Global Prime - 4 times as many ticks as... FXCM (standard account)
The price will go up, and down, not necessarily in that order.
  • #117
  • Apr 14, 2016 9:45am Apr 14, 2016 9:45am
Quoting ForexFart
Global Prime - 4 times as many ticks as... FXCM (standard account)
So you divided the global prime x-axis by 4 and then superimposed the two graphs? On set of bid/ask lines does seem to have more jaggedness to it, and I suppose that would be the global prime feed.
  • #118
  • Apr 14, 2016 7:06pm Apr 14, 2016 7:06pm
Well, I can't just rescale that easily. Ticks don't have a timescale. There might be double the amount of ticks in 1 minute but the same amount of ticks in another. So I divided the global prime image up into 10 seconds and rescaled each section over the fxcm image. But that's just technicalities that are interesting but not overly relevant. The point is, they're both of the same scale now (well, close enough) to overlay on each other.

Yes, Global Prime has more movement on bid / ask which makes sense given the extra ticks that come though.

The other thing to consider here is that FXCM's bid price (on the chart) actually isn't accurate (and I think neither is the ask). The prices in the "market watch" are correct, but the chart is not. It's an INSANE concept to me which FXCM has been promising to fix for month (years?), but I don't want to skew this thread off into a whole discussion about that. The point is my tick collector is using the market watch prices.
The price will go up, and down, not necessarily in that order.
  • #119
  • May 7, 2016 11:17am May 7, 2016 11:17am
when the indicator option showcumulative is set to true, it accumulates from which point on ? can somebody please tell me....or if I need to "accumulate the delta" for a particular number of price bars...then how to calculate...
somebody pls help....
  • #120
  • May 7, 2016 11:54am May 7, 2016 11:54am
Quoting joshychris
when the indicator option showcumulative is set to true, it accumulates from which point on ? can somebody please tell me....or if I need to "accumulate the delta" for a particular number of price bars...then how to calculate... somebody pls help....
Hi JC. It will begin to accumulate the moment you attach it to the chart. The difference in showcumulative set to true is in how the data will be displayed. With it set to true, the opening of the bar will begin at the close of the prior bar. With it set to false, the opening of the bar will always be zero.
0 traders viewing now