Support Board
Date/Time: Wed, 27 Nov 2024 06:33:06 +0000
VWAP Difference
View Count: 4560
| [2014-04-30 08:09:21] |
| Dom7364 - Posts: 257 |
|
Why is there a different VWAP if on a 4 tick range bar for example and just a 1 tick chart pls? Thanks |
| [2014-04-30 08:22:49] |
| |
|
Refer to the documentation here about this study: http://www.sierrachart.com/index.php?l=doc/doc_TechnicalStudiesReference.html#s108 Refer to the Base on Underlying Data input. Sierra Chart Support - Engineering Level Your definitive source for support. Other responses are from users. Try to keep your questions brief and to the point. Be aware of support policy: https://www.sierrachart.com/index.php?l=PostingInformation.php#GeneralInformation For the most reliable, advanced, and zero cost futures order routing, *change* to the Teton service: Sierra Chart Teton Futures Order Routing |
| [2014-04-30 09:16:12] |
| Dom7364 - Posts: 257 |
|
I have, been using it for quite a while, and there is a difference, albeit slight, if looking at a 1 tick data or another time period (2 ticks, 2 mn, 30 mn, etc). I just wanted to understand why.
|
| [2014-04-30 09:19:03] |
| |
|
Yes, there will be a difference if you are not using the Base on Underlying Data option.
Sierra Chart Support - Engineering Level Your definitive source for support. Other responses are from users. Try to keep your questions brief and to the point. Be aware of support policy: https://www.sierrachart.com/index.php?l=PostingInformation.php#GeneralInformation For the most reliable, advanced, and zero cost futures order routing, *change* to the Teton service: Sierra Chart Teton Futures Order Routing |
| [2014-05-07 17:42:19] |
| User33737 - Posts: 8 |
|
I thought this was fixed a while ago by using the underlying data opion. If I use a 1 tick chart and overlay VWAP on another time frame, the differences can be 4-5 points. But if I use a 1 second chart they are virtually the same. I am thinking the underlying data being used to calculate vwap is a 1 second OHLC? maybe that would clear this up for me. Thanks again! |
| [2014-05-07 18:06:35] |
| |
|
Why do you overlay the Volume Weighted Average Price study from one chart onto another chart? How do you have the Base on Underlying Data input for the study set? Sierra Chart Support - Engineering Level Your definitive source for support. Other responses are from users. Try to keep your questions brief and to the point. Be aware of support policy: https://www.sierrachart.com/index.php?l=PostingInformation.php#GeneralInformation For the most reliable, advanced, and zero cost futures order routing, *change* to the Teton service: Sierra Chart Teton Futures Order Routing |
| [2014-05-07 18:55:46] |
| User33737 - Posts: 8 |
|
Merely to see the difference between. I don't overlay for real use. I set to yes for underlying data. |
| [2014-05-07 19:35:34] |
| |
|
On the ES futures, comparing between 1 Number of Trades Per Bar and 1 minute, we see a difference of .02 for the Volume Weighted-average Price for the trading day. This is very tiny. So the calculations are virtually the same. At higher time frames per bar, there are no differences comparing between different time frames. We do not know exactly what you are seeing, the exact settings of the study and exactly what you are doing, so we cannot comment specifically on your questions because they are too vague. It is completely impossible for us to help you with a small amount of information given. We would have to get a simple Chartbook from you and test it and we will not do that because we do not see any problems with this study. And we will not help with differences you see when overlaying this study on another chart. There should not be a problem with this. But we do not know exactly what you are doing, the settings and why you think there is a difference Sierra Chart Support - Engineering Level Your definitive source for support. Other responses are from users. Try to keep your questions brief and to the point. Be aware of support policy: https://www.sierrachart.com/index.php?l=PostingInformation.php#GeneralInformation For the most reliable, advanced, and zero cost futures order routing, *change* to the Teton service: Sierra Chart Teton Futures Order Routing Date Time Of Last Edit: 2014-05-07 19:38:51
|
| [2014-05-07 20:17:29] |
| |
|
Here are the images from the testing we did: http://www.sierrachart.com/image.php?l=1399493555234.png http://www.sierrachart.com/image.php?l=139949383195.png Sierra Chart Support - Engineering Level Your definitive source for support. Other responses are from users. Try to keep your questions brief and to the point. Be aware of support policy: https://www.sierrachart.com/index.php?l=PostingInformation.php#GeneralInformation For the most reliable, advanced, and zero cost futures order routing, *change* to the Teton service: Sierra Chart Teton Futures Order Routing |
| [2014-05-07 21:43:25] |
| User33737 - Posts: 8 |
|
Here is a chartbook with a 1 tick and 5 minute ES. I may be completely wrong, but I think at 10:30cst on this sample the 1 tick vwap is 1866 and the 5 min is 1863. The data is from CTS. thanks for looking into this. Date Time Of Last Edit: 2014-05-07 21:43:52
|
 vwap-test.cht - Attached On 2014-05-07 21:39:03 UTC - Size: 57.76 KB - 533 views |
| [2014-05-08 04:45:57] |
| |
|
We do not need the Chartbook based upon further testing. We have shown that there will be no differences when changing the timeframe per bar when basing on the underlying data. We tested with the default study settings other than basing on the underlying data. Change the timeframe per bar within the same chart. You should not notice any differences of significance. If you do, re-download the data with Edit >> Delete All Data and Download. Sierra Chart Support - Engineering Level Your definitive source for support. Other responses are from users. Try to keep your questions brief and to the point. Be aware of support policy: https://www.sierrachart.com/index.php?l=PostingInformation.php#GeneralInformation For the most reliable, advanced, and zero cost futures order routing, *change* to the Teton service: Sierra Chart Teton Futures Order Routing Date Time Of Last Edit: 2014-05-08 07:25:32
|
| [2014-07-25 00:45:58] |
| User82828 - Posts: 44 |
|
I have also differences for the VWAP and 1st Standard deviation high/low when trading the dax future. I use IQfeed as datafeed. So if I compare a 5-min and 4-min chart, the VWAP and the upper and lower 1st Standard deviation is different on both charts by up to 1-6 ticks. The value for "Based on Underlying Data" is set to "No", which is the default, so then the price and volume of the calculations is based on the bars in the chart according to your technical referenence at your homepage. With this information I can a little bit understand the difference, despite the fact that the VWAP and standard deviation are variables, which do not depend on time and should therefore generally not depend and differ on the selected 4min or 5 min bartime or chart. Then I changed the settings for "Based on Underlying Data" to "yes" so that the time period type and Time Period lenght is ignored according to your technicsl reference on the SC homepage. However, today the dax showed in the 5min Chart and 4min Chart with these settings also in the first hour and later a difference of up to 6 ticks for 1st Standard deviation (std. Deviation Distance Multiplayer was set to "1"). How can this be? Which Value for the VWAP and standard deviations are correct and which settings do you recommend. Why is there a difference even if use the "yes" settings for "Based on Underlying Data"? Generally the VWAP and standard Deviation are not depending on any time settings, so in general, they should be equal on all time frames for one day, right? |
| [2014-07-25 09:45:50] |
| User30668 - Posts: 108 |
|
User82828
Generally the VWAP and standard Deviation are not depending on any time settings, so in general, they should be equal on all time frames for one day, right?
This is correct, the VWAP and its Standard Deviations should show always the same value in every time frame.Nevertheless the very same issue is happening to me too. I use several time frames charts (1tick, 1min, 5 min, 30min) and frequently (not always) there is a considerable difference between the Standard Deviation values from chart to chart. And I am referring to real time action. I have the "Based on Underlying Data" set to "yes" in all VWAP studies, as advised by SC Engineering. As a workaround, besides the Vwap study on each chart I also have a Study/Price Overlay from the 5min chart to the other charts, to be able to recognize the differences. In any case, when there are differences I never know which one is right or wrong. It would be much better if we could just rely on the VWAP study values on every different time frame or at least know for sure which time frame is showing the correct values, as these divergences can sometimes lead to incorrect entry levels. So SC Engineering, would you please put this issue on your big list of tasks and see what you can do about it when / if possible? It would be much appreciated. Thank you. Date Time Of Last Edit: 2014-07-25 09:51:01
|
| [2014-07-25 10:15:22] |
| User82828 - Posts: 44 |
|
Yes, I 100% agree with User30668's statement and made the same experience in all details, he described. For swing trading positions, the differences in VWAP and the corresponding Standard deviation of up to 6 ticks/points might be just still ok. However for scalping setups with tiny stop losses and with first profit targets of e.g. 6 ticks/points, it is not acceptable to have an inaccuracy at this same level, because this could lead often to incorrect entry levels with corresponding potential losses. So I would also very much apreciate, if SC Engineering could answer and comment these last postings and fix this bug soon. Thank you very much. Date Time Of Last Edit: 2014-07-25 10:16:20
|
| [2014-07-25 17:45:18] |
| |
|
We will look into this. We realize that when using the option to base the Volume Weighted Average Price on the underlying volume at price data, there could be a problem in the way that the calculations are done when the chart is updating in real-time. Do the differences go away between charts when you select Chart >> Recalculate? So if I compare a 5-min and 4-min chart, the VWAP and the upper and lower 1st Standard deviation is different on both charts by up to 1-6 ticks. The value for "Based on Underlying Data" is set to "No", which is the default, In this particular case, there will be differences and that can be expected.
Sierra Chart Support - Engineering Level Your definitive source for support. Other responses are from users. Try to keep your questions brief and to the point. Be aware of support policy: https://www.sierrachart.com/index.php?l=PostingInformation.php#GeneralInformation For the most reliable, advanced, and zero cost futures order routing, *change* to the Teton service: Sierra Chart Teton Futures Order Routing Date Time Of Last Edit: 2014-07-25 17:46:07
|
| [2014-07-28 10:32:27] |
| User82828 - Posts: 44 |
|
I think we have to clarify and discuss the VWAP-Price Difference on a general level: VWAP is an important institional observed parameter which is recalcualted with every new trade over the entire day. The calculation of VWAP works in this way: For each trade, the Price is multiplicated with the number of contracts and this results will be added and added over the day for every new trade. Simultaneously this multiplied and added values are divided by the sum of all traded contracts, which finally results in the exact calulation of ONE VWAP-value. This proofs that VWAP ist therefore the average Price of all traded contracts and does not depend on any timeframe or any lenght of a bar chart, etc... Taking into acount this cirumstances, this also proofs, that there can only be ONE true VWAP-Value at any time of the day. If I have at the same time different VWAP values (as we now have in SC), this definitely means a wrong and/or inaccurate calulation. So my strong recommendation would be, that the VWAP calculation is completely overworked within SierraChart, resulting in a calculation of the one and correct VWAP value, independent of the used timeframe or calualtion method, one is using. Please correct me, if I am wrong. I forgot to mention, that 100% correct, accurate and complete tick data are required for such a VWAP calualtion. Therefore I use the iqfeed datafeed where all tick data come directly from the exchange e.g for DOW, S&P or DAX, etc. I would appreciate a comment from SC support to this posting. Date Time Of Last Edit: 2014-08-02 08:03:42
|
| [2014-07-28 13:59:17] |
| User30668 - Posts: 108 |
|
Sierra Chart Engineering
Do the differences go away between charts when you select Chart >> Recalculate?
No, the differences remain the same after recalculation.SC Engineering, I have been running some tests in order to get together some more information that I hope might help to fix this issue. Regarding the time based charts, the divergences seem to be more evident just on the Deviation Bands. The Vwap value itself usually shows the same value or minimum differences. I was able to reproduce the issue with SC Replay and made a 2 minutes video showing the divergence problem. The video is showing the following: - On the left side of the screen there is a 5min chart with a Vwap study; - On the rest of the screen there is a 1 min chart also with a Vwap study and a Study/Price overlay based on the Vwap study that is running on the 5min chart (just to make it easier to compare); - Both Vwap studies are set to "Base On Underlying Data" and to show the Vwap value and Deviation Bands 4. The replay is from July 7th, from 9:07 am US EST (in case you want to replay it your self). I apologize if the file is a bit too big. I hope this information helps you to better understand this Vwap "issue". I trust you will find the best solution for it. Thank you. |
| Vwap Deviation Band 4 divergence.avi - Attached On 2014-07-28 13:24:40 UTC - Size: 13.72 MB - 550 views |
| [2014-07-30 15:21:29] |
| Al SC Developer - Posts: 434 |
|
The VWAP is a fairly simple calculation, but is very dependent on the data that it is calculated over. When you use the underlying data, you are effectively using the data that makes up the bars. This can still not be perfectly accurate though if you do not have tick data, because there will be slight variations (i.e. 1-sec bar can only estimate the actual ticks that make up the underlying bar which can result in slight skews). When you compare VWAPs on different time frames, the values should be very close if you are using underlying data. If you are not using underlying data, then the values will be different. The standard deviations from different time frames will normally be different. This is because the standard deviation is calculated over the bar values of the VWAP, and you have vastly different values (i.e. you only have every fifth value on a 5-min chart vs a 1-min chart). |
| [2014-08-02 03:21:55] |
| |
|
The information from post # 18 has been added to the documentation now.
Sierra Chart Support - Engineering Level Your definitive source for support. Other responses are from users. Try to keep your questions brief and to the point. Be aware of support policy: https://www.sierrachart.com/index.php?l=PostingInformation.php#GeneralInformation For the most reliable, advanced, and zero cost futures order routing, *change* to the Teton service: Sierra Chart Teton Futures Order Routing |
To post a message in this thread, you need to log in with your Sierra Chart account: