What was ES Value Area on the 2010-11-01 (YYYYMMDD)

November 2nd, 2010, 10:51 AM

LukeGeniol

@ trendisyourfriend, don't know what indicator do u use on NT, but I think the difference u get is due to the difference on how the indicator make the calc. IRT and MD use tick, NT indicators usually use the same timeframe of chart where u use them, so is an approximation, and lower is timeframe more is accurate. U maybe can try Fin-Alg market profile that is able to use tick for calcs.

Thanks Luke, i tried Fin-Alg and it suffers from the same problems. Also, it's less powerfull than what we can find in I/RT or MD. What bugs me is that i thought that all TPO profile should give the same values. This is not the case and i can't find a logical explanation.

November 2nd, 2010, 11:00 AM

trendisyourfriend

Thanks Luke, i tried Fin-Alg and it suffers from the same problems. Also, it's less powerfull than what we can find in I/RT or MD. What bugs me is that i thought that all TPO profile should give the same values. This is not the case and i can't find a logical explanation.

We just need to explore this. Market Delta, as stated below, uses 70% of the TPOs. For one standard deviation you would want to take 68.2% of the TPOs. Thta is already a small difference.

https://www.marketdelta.com/xm_client/client_documents/80%20Percent%20Rule%20Strategy.pdf

Again the result depends on the resolution. 15 min TPOs will give you a different result than 30 min TPOs. The thing will only become accurate, if you use tick data and a volume distribution that has been generated independently from the bars or candle sticks.

The market delta TPO indicator has a specific setting for tick increments, for ES you can set that to 0.25. Not sure that this means that the indicator really uses tick data. It probably only divides the bar length by the tick increment to create the TPOs.

https://support.marketdelta.com/entries/188635

November 2nd, 2010, 11:10 AM

trendisyourfriend

Thanks Luke, i tried Fin-Alg and it suffers from the same problems. Also, it's less powerfull than what we can find in I/RT or MD. What bugs me is that i thought that all TPO profile should give the same values. This is not the case and i can't find a logical explanation.

Dont think, this is how works TPO from dValueArea indicator:

Code

TPO - This method disregards volume altogether, and gives a single hit to each price in the range of the bar.
    //                        e.g.  A 5-minute bar has a range of 1.5 points with a High at 1534 and a Low at 1532.5, then
    //                        1 contract (or "hit") is loaded to the seven prices in that range: 1532.50, 1532.75, 1533.0, 1533.25, 1533.50, 1533.75, and 1534

So lower timeframe more bars, more range......

November 2nd, 2010, 11:17 AM

MarketDelta already uses to different methods to calculate the value area. Method A is also supported by E-Signal, method B is also used by CQG. Details explained below.

Of course, both MarketDelta method will produce different results. Which one shall we use?

MarketDelta Support : How Is The [AUTOLINK]Value Area[/AUTOLINK] Calculated for the Footprint and Market Profile Charts?

November 2nd, 2010, 11:25 AM

Fat Tails

We just need to explore this. Market Delta, as stated below, uses 70% of the TPOs. For one standard deviation you would want to take 68.2% of the TPOs. Thta is already a small difference.

https://www.marketdelta.com/xm_client/client_documents/80%20Percent%20Rule%20Strategy.pdf

The market delta TPO indicator has a specific setting for tick increments, for ES you can set that to 0.25. Not sure that this means that the indicator really uses tick data. It probably only divides the bar length by the tick increment to create the TPOs.

MarketDelta Support : [AUTOLINK]TPO[/AUTOLINK] Indicator

Fat Tails

MarketDelta already uses to different methods to calculate the value area. Method A is also supported by E-Signal, method B is also used by CQG. Details explained below.

Of course, both MarketDelta method will produce different results. Which one shall we use?

MarketDelta Support : How Is The [AUTOLINK]Value Area[/AUTOLINK] Calculated for the Footprint and Market Profile Charts?

Anyway on MD the POC VAH and VAL of TPO indicator are the same on every timeframe.

November 2nd, 2010, 11:35 AM

Fat Tails

MarketDelta already uses to different methods to calculate the value area. Method A is also supported by E-Signal, method B is also used by CQG. Details explained below.

Of course, both MarketDelta method will produce different results. Which one shall we use?

MarketDelta Support : How Is The [AUTOLINK]Value Area[/AUTOLINK] Calculated for the Footprint and Market Profile Charts?

The indicators available presently for Ninja should use method A or B no matter as the differences should not be as significant as what i posted in the first comment. As of now i consider Ninja has no serious market profile available making the platform much less attractive in my eyes.

November 2nd, 2010, 11:44 AM

LukeGeniol

Anyway on MD the POC VAH and VAL of TPO indicator are the same on every timeframe.

That's how it should be otherwise it would not make sense. Only on Ninja do we find such idiosyncrasy .

November 2nd, 2010, 11:55 AM

trendisyourfriend

The indicators available presently for Ninja should use method A or B no matter as the differences should not be as significant as what i posted in the first comment. As of now i consider Ninja has no serious market profile available making the platform much less attractive in my eyes.

The MarketDelta value area definitions are somewhat arbitrary. Let us imagine that you have a vertical volume distribution with three peaks. Highest peak in the middle is POC. Now the other peaks are separated from the main peak by valleys. If one of the valley is deeper than the other one, the price action on the other side of the deeper valley will be totally excluded, even if the peak there is higher.

On the other side, if the MarketDelta results are independant from the time frame chosen, they may use a very high resolution, suhc as tick data.

Does MarketDelta show the same results for the value area on a 1 min chart, a 30 min chart, a 5 range chart and a 277 tick chart?

November 2nd, 2010, 12:16 PM

Fat Tails

The MarketDelta value area definitions are somewhat arbitrary. Let us imagine that you have a vertical volume distribution with three peaks. Highest peak in the middle is POC. Now the other peaks are separated from the main peak by valleys. If one of the valley is deeper than the other one, the price action on the other side of the deeper valley will be totally excluded, even if the peak there is higher.

On the other side, if the MarketDelta results are independant from the time frame chosen, they may use a very high resolution, suhc as tick data.

Does MarketDelta show the same results for the value area on a 1 min chart, a 30 min chart, a 5 range chart and a 277 tick chart?

Your question is similar to asking if 1200 price level on a 1 min chart is more accurate than 1200 price level on a 30min chart. If we use a MP based on TPO's the standard convention is to use a 30min chart. Everyone else use that interval except those who prefer transcendental numbers. But more seriously, if you use a 1 min chart you'd still want to know the numbers coming from the standard 30min interval. That's how i look at MP as an absolute area developing during the day.

EDIT: i have checked on I/RT and whether or not we change the interval the value area stays the same.

November 2nd, 2010, 04:47 PM

trendisyourfriend

The indicators available presently for Ninja should use method A or B no matter as the differences should not be as significant as what i posted in the first comment. As of now i consider Ninja has no serious market profile available making the platform much less attractive in my eyes.

I don't know how do u use it and if u find it useful, but in my view MP is an old good tool for the time steidlmayer invented it, but not more actual for today markets. Now there are more powerfull tools, starting from volume profile, vwap with standard deviation......etc...ie statistics applied to volumes.

What was ES Value Area on the 2010-11-01 (YYYYMMDD)

Discussion in Emini and Emicro Index

What was ES Value Area on the 2010-11-01 (YYYYMMDD)