Support Board
Date/Time: Mon, 25 Nov 2024 06:46:42 +0000
[User Discussion] - Differences in VbP
View Count: 2693
| [2014-01-06 00:25:20] |
| PTG - Posts: 57 |
|
I notice differences in volume profiles for 6E between SC feed and IQFeed: http://www.sierrachart.com/image.php?l=1388966880899.png (SC Feed 6EH4 continuous back adjusted volume based, no difference with date rule based) http://www.sierrachart.com/image.php?l=1388966818395.png (IQFeed EU#C continuous back adjusted). The settings for both are the same for VbP as well as the data configuration (tick by tick). The data from the SC feed was freshly loaded, the data from IQFeed is from Dec 31st. I cannot reload it as I have cancelled my subscription. That said, I haven't noticed data anomalies in the past fwiw. The question now is which one is an accurate representation of reality. Please advise. |
| [2014-01-08 00:13:21] |
| |
|
One question we have is whether the data from IQ feed is actually tick by tick. During market hours they have a limitation of 8 days of tick data. So you could have some minute data. The other thing is you need to be certain certain that the rollover point is exactly identical. So you are saying you do not notice any difference with date-based rollover? Also, the settings are different between these two profiles. That is clear. Really this question is getting outside the scope of our support. We are not really sure what the purpose of the question is. You are in a better position to really know what's going on because you created the profiles. We are at a disadvantage. 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 |
| [2014-01-08 01:52:26] |
| PTG - Posts: 57 |
|
"One question we have is whether the data from IQ feed is actually tick by tick.
That's a fair point. The .scid file I downloaded is dated 1/1/2014 12:39 AM (GMT+1). The size is 82.589 kB. It starts July 10th 2008. How can I check if the file is actually tick by tick ? The other thing is you need to be certain certain that the rollover point is exactly identical.
I have no control over EU#C. It is the backadjusted contract from IQFeed. They roll over according to their rules here: the Normal continuous and Back-adjusted Continuous Futures follow the rules that are shown on our Symbol Guide at http://iqfeed.net/symbolguide/index.cfm?symbolguide=guide&displayaction=support§ion=guide&web=iqfeed. So you are saying you do not notice any difference with date-based rollover?
I'm saying that for that particular profile there is no difference whatsoever between date based and volume based rollover on the SC datafeed. Also, the settings are different between these two profiles. That is clear.
One profile is from IQFeed, the other from SC. The IQFeed continuous contract is set to None because it is continuous already, right ? We are not really sure what the purpose of the question is.
I want to know of any differences between IQFeed, which I have been using for years, and the SC feed. I do realize that there are initial issues (as reflected in other threads as well), and so should SC imho as a supplier. After all, the SC feed isn't identical to IQFeed, it's better I'd hope. So far, the switch to SC feed has cost me extra time. You are in a better position to really know what's going on because you created the profiles. We are at a disadvantage.
I'm kind of in a void since I cannot reload the IQFeed data. I may have switched to the SC datafeed too soon. Maybe I'm the one who's disadvantaged here, what do you think ? Let's first find out if the file I have is tick by tick, because if it isn't, then that could explain the difference. Let me know please. Date Time Of Last Edit: 2014-01-08 02:24:43
|
| [2014-01-08 03:29:17] |
| |
|
One profile is from IQFeed, the other from SC. The IQFeed continuous contract is set to None because it is continuous already, right ? Inevitably, there will be some differences between different data feeds for various reasons. The data the Sierra Chart futures data feed provides is good data. It does not mean that it is 100 % perfect. It is 99.9999% perfect. IQ Feed data is going to be less perfect because they to not have as much tick by tick history and place limitations on it. We do not even know how far back these profiles you are creating even go or what the time span is. In the end, we have to go through a time-consuming, painstaking, difficult analysis and then provide explanations as to what is going on with the particular scenarios you have developed. You would have to provide your chart book and your data files. In the end, it does not change anything. This is why this is outside the scope of our support. This is a matter of policy for good reason. All we can say is the Continuous Futures Contract feature in the latest Sierra Chart versions works very well. The Sierra Chart futures data feed provides tick by tick data for the last three years. This is a lot more than IQ Feed. As long as you have a tick by tick data configuration: http://www.sierrachart.com/index.php?l=doc/doc_TickbyTickDataConfiguration.php you are going to have accurate profiles. This is all we can say. We cannot spend any further time on this. Some of the issues we had in the beginning, with the data feed, would cause microscopic and virtually nondetectable inaccuracies with volume profiles during that time. 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: 2014-01-08 03:30:26
|
| [2014-01-08 03:44:59] |
| |
|
We have also added an additional historical data for the following contracts: 6A, 6B, GC. When it comes to adding more data, we are happy to do that. This is going to be 1 minute data. We increased it to 4 years but we may be able to go more. 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 |
| [2014-01-08 04:18:32] |
| |
|
One thing we observed in your chart images is that the IQ feed volume profiles have a smoother appearance on the right side. We wondered what this meant. If you have a look at the below images, you can see that 1 Minute data has a smoother appearance on the volume profiles compared to the 1 tick data. Minute data: http://www.sierrachart.com/image.php?l=1389154101693.png 1 Tick data: http://www.sierrachart.com/image.php?l=1389154202956.png What this means, is that the Sierra Chart Volume Profiles are far more accurate than what you have with IQ Feed. There simply is no comparison here. The market data platform that we have developed, is on a much higher standard than IQ Feed. Yes, there can be small issues in the historical data, mainly missing ticks. But you can drive yourself absolutely crazy trying to perfect every one of these little defects. You need to understand our perspective on this. The data we have is 99.9999% accurate. If you are on the other side of this Support Board, answering questions, you may have a tremendous appreciation at how calm and cool we actually are here. 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: 2014-01-08 04:20:27
|
| [2014-01-08 13:05:06] |
| PTG - Posts: 57 |
|
If you are on the other side of this Support Board, answering questions, you may have a tremendous appreciation at how calm and cool we actually are here.
LOL. I'm not on your side of this SB but I can imagine what is going on very well. It is extremely difficult to communicate about issues. You are dealing with individuals with various competence levels in unique circumstances. One thing we observed in your chart images is that the IQ feed volume profiles have a smoother appearance on the right side. We wondered what this meant. If you have a look at the below images, you can see that 1 Minute data has a smoother appearance on the volume profiles compared to the 1 tick data.
Yes, I see that. They are from ES though, not 6E. I wonder why they aren't identical. I'd suggest to check the VbP settings (no pun intended ;) ) I have set the Intraday Data Storage Time Unit to 1 minute, and performed a Delete All Data And Download. The resulting profile is the following: http://www.sierrachart.com/image.php?l=1389181375573.png This is SC feed data. I then have set the Intraday Data Storage Time Unit back to 1 tick, and performed a Delete All Data And Download once more. The resulting profile is the following: http://www.sierrachart.com/image.php?l=1389181863993.png This is SC feed data. There is no difference. I'd expect this since the VbP would only differ if the time unit considered were smaller than 1 minute, which is not the case. As I see it, it makes no difference if you add up VbP over a period of days based on intraday data records containing tick data vs. minute data, as long as the period is a multiple of minutes. Note that the file size for 6EH4.scid was 44.676 Kb for tick data vs. 728 Kb for minute data so the delete and download data did actually take place. Please correct me if I'm seeing it wrong. We are referring to the Volume by Price settings. They are different in your charts.
Most definitely not: they are the same. 100%. I'm afraid all this is much simpler. The difference is in the .scid file from IQFeed on the computer I made the original screenshot from. After performing the above, I suspected the IQFeed based .scid file. I found another .scid file for IQFeed's @EU#C on another computer, fortunately. I hadn't thought of that since it wasn't supposed to be there. The screenshot for the profile based on that IQFeed .scid file is this: http://www.sierrachart.com/image.php?l=1389183997321.png Compare this with any of the two above (both are based on SC data), and the differences are rather small, something that I'd expect. I have no idea why the .scid file on the other computer is off. It shouldn't be, but it is what it is. This appears to be the cause for the difference. All in all I have my answer. I'm confident about the SC data. That said, I don't understand why I see identical profiles for tick- and minute data on 6E where you see different profiles on ES. |
| [2014-01-11 19:45:49] |
| |
|
There is no difference. I'd expect this since the VbP would only differ if the time unit considered were smaller than 1 minute, which is not the case. As I see it, it makes no difference if you add up VbP over a period of days based on intraday data records containing tick data vs. minute data, as long as the period is a multiple of minutes. Note that the file size for 6EH4.scid was 44.676 Kb for tick data vs. 728 Kb for minute data so the delete and download data did actually take place. Please correct me if I'm seeing it wrong.
There must be a difference. There is 0% chance they would be the same. We understand exactly how this works and there will be a difference for sure. In our own test we have proved that. Since we do not know exactly what you have done in your test, we cannot comment on your results. Most definitely not: they are the same. 100%
Look at the top line of the chart images you have given in post #1. You can clearly see that the input settings for each of the Volume by Price studies are different. If those input settings we are seeing are for another volume profile that we cannot see, then you have overly complicated the image and have confused things. This is what causes such a problem for us. 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 |
| [2014-01-12 00:10:45] |
| PTG - Posts: 57 |
|
Look at the top line of the chart images you have given in post #1.
It's not about the top lines. It's about the actual settings for the volume profiles displayed. I've been talking about the profile starting 0800 on Nov 1st 2013 all the time. How can you conclude the settings for SC and IQ are different based on the top line ? Perhaps it is the fact that I have more profiles on the chart which I have hidden ? We understand exactly how this works
I hope so. |
| [2014-01-12 00:47:19] |
| PTG - Posts: 57 |
|
The study settings for the profiles displayed (F6 and select the applicable profile) are the same for the charts displayed. There is a 0% chance they are different. I'm perfectly capable to compare two F6 - Settings - Settings and Input and Subgraphs and Alerts. It's not that difficult you know. I'm not retarded, or so I hope. I'm happy to send you screenshots of the settings - they are the same. 100% the same. As in: identical. The only thing different is the data loaded - 1 minute vs. 1 tick data. Between http://www.sierrachart.com/image.php?l=1389181863993.png and http://www.sierrachart.com/image.php?l=1389181375573.png there is no difference, once again. I have many profiles on my charts. The top line is probably not indicative. If I tell you the settings are the same for a particular profile, you can believe me on my word. I am not wasting my time on BS, or am I. There is no reason to suppose I am overcomplicating things. Date Time Of Last Edit: 2014-01-12 00:47:48
|
| [2014-01-12 23:37:54] |
| PTG - Posts: 57 |
|
Updated to V1077 on another PC (PC1) Created a chartbook. Connected to TransAct Simulated Live and Sierra Chart datafeed. Updated to V1077 on my laptop (PC2). Created a chartbook. Connected to TransAct Simulated Live and Sierra Chart datafeed. IDSTU = Intraday Storage Time Unit. The other relevant settings on both PCs are identical, including Days to Load (300). Once again, the profile settings for this profile, which is the only profile on the chart, are exactly the same, on both PC's. Between Screenshot 1/2 and 3/4 the data has been deleted by the Delete All Data and Download command. Before Screenshot 1 and 3 the data was also deleted and downloaded. All this means that fresh data has been downloaded. Screenshot 1: 1 tick IDSTU on PC1 http://screencast.com/t/cyrLjyx8j Screenshot 2: 1 Minute IDSTU on PC1 http://screencast.com/t/ePm5mdJELzb Screenshot 3: 1 tick IDSTU on PC2 http://screencast.com/t/ux3qMRADz Screenshot 4: 1 Minute IDSTU on PC2 http://screencast.com/t/JC9ggZIPgbq So: what do I see ? 1. Screenshot 1 and 2 are identical. Yet there is different data: 1 tick vs. 1 min. 2. Screenshot 3 and 4 are identical. Yet there is different data: 1 tick vs. 1 min. 3. there is a difference between 1/2 on one hand and 3/4 on the other. Not a lot, but there is difference. 4. the [M] on Screenshot 1 and 2 isn't there. This is because I can only connect to one SC data feed at a time. The screenshot was taken after I had connected the other SC on the laptop. Settings screenshots: http://screencast.com/t/yi3s2hrkALOZ http://screencast.com/t/0CT6g6RpA6 http://screencast.com/t/l2ZO89W9QO http://screencast.com/t/0ceI4utYE I trust this makes it less complicated. If not, let me know what you require. Date Time Of Last Edit: 2014-01-12 23:39:59
|
To post a message in this thread, you need to log in with your Sierra Chart account: