Home >> Support Board

Support Board

Date/Time: Thu, 25 Apr 2024 23:47:58 +0000

How to Post a Message for Support

Data deivates between sub-instances

View Count: 457

[2022-01-12 13:53:32] #1 [Go To First Post]
jwick - Posts: 142
I'm using several sub-instances to backtest my strategies. I'm noticing the actual market data sometimes deviates (different file sizes) between sub-instances. My main instance uses 1 tick resolution while all the sub-instances use 1 second resolution. The sub-instances are all using the same settings and chartbooks. Any ideas why this might be occurring? I'm pulling in futures data from 2008 till the present and do periodically roll over my front month contract as time moves forward (not sure if the data begins to deviate when I perform that rollover procedure). I have attached an example of one such file. Date Time Of Last Edit: 2022-01-12 14:11:29


Private File Private File

[2022-01-12 14:42:45] #2 [Go To First Post]
Sierra Chart Engineering - Posts: 104368
We cannot analyze this. We do not know the specific reason. As long as the charts look the same between instances that is the important thing. 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: 2022-01-12 14:43:08

[2022-01-12 15:51:02] #3 [Go To First Post]
jwick - Posts: 142
The underlying charts are difference. My strategy produces the same results in a chart replay within one instance but slightly different results across instances. I also verified this by manually copying all of the scid files from one instance into another and the strategy now produced the same results on both instances. Date Time Of Last Edit: 2022-01-12 15:51:27

[2022-01-12 15:54:25] #4 [Go To First Post]
Sierra Chart Engineering - Posts: 104368
Are the instances receiving delayed data? Not sure if this is related but just checking. 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

[2022-01-12 16:29:45] #5 [Go To First Post]
jwick - Posts: 142
I believe they aren't receiving delayed data. The main instance isn't receiving delayed data, so I'm assuming the sub-instances don't either. To add some details on the chart setup, these are on index futures (ES, NQ, YM). The end date on all the charts is always in the past (currently 12-2021) so it shouldn't be streaming in more data in real-time. I use the sub-instances to test my strategy (chart replay) against historical data and when there is a rollover in the actual front-month, I add a few months to the end of every chart (and update the front month on every chart (continuous back adjusted)). So, for example, sometime in mid-March / April of this year after the rollover, I'll update the end date on the charts to 02-2022 and update the front month to 2022-06. I think I experienced something similar and around this time of switching the front month on the sub-instances + adding a few months to the end of every chart, but I'm honestly not sure if that was actually when I encountered this before. EDIT: Additional maybe relevant data ... all sub-instances are downloading and storing volume at price information. Not sure if that is causing the discrepancy. Date Time Of Last Edit: 2022-01-12 16:43:24

[2022-01-12 16:55:54] #6 [Go To First Post]
jwick - Posts: 142
hmm ... this is interesting. So, this is what I did: 1. Deleted all dly and scid files in sub instance A. Opened chartbook and let data "download" into the instance 2. Deleted all dly and scid files in sub instance B. Copied dly and scid files from sub instance A to B. Opened chartbook and connected to data feed Now there are indeed differences between the dly and scid files between the two subinstances. Specifically (if that matters): file idA idB sizeA sizeB year 4 NQ-201406-GLOBEX.scid 5 8 20290296 56206296 201406 0 ES-202203-GLOBEX.scid 5 8 43197136 43209976 202203 1 MES-202203-GLOBEX-USD.scid 5 8 41025216 41037896 202203 2 MNQ-202203-GLOBEX-USD.scid 5 8 55931016 55944296 202203 3 MYM-202203-ECBOT-USD.scid 5 8 24437256 24446656 202203 5 NQ-202203-GLOBEX.scid 5 8 44187576 44200776 202203 6 YM-202203-ECBOT.scid 5 8 27979776 27990416 202203 I'm running a chart replay to observe if the strategy results differ as well. EDIT: Please ignore above for the moment. I'm verifying this one more time with two other instances. It is possible that while I did the copy of dly and scid files from A to B, I had started the chart replay on A (and I'm not sure if that effects SCID files). Currently "downloading" a fresh copy of data into another sub-instance. Date Time Of Last Edit: 2022-01-12 17:38:10

To post a message in this thread, you need to log in with your Sierra Chart account: