Support Board
Support Board
Search Board
Control Panel
View My Posts / Threads
Direct MessagesDate/Time: Sun, 28 Apr 2024 20:34:16 +0000
Problem with volume-based rollover
View Count: 1999
| [2016-12-01 04:10:27] |
| skelcap - Posts: 139 |
|
I redownloaded the data but nothing changed. Your volume-based rollover algorithm is capturing all of the other roles in COIL reasonably well, so looks like it's just some kind of glitch with those two contracts CBK11 and CBX15.
|
| [2016-12-01 04:24:46] |
| |
|
We think the problem is because the rollover rules were different back at that time. We are not optimistic about being able to solve this. But we will look at it when we have a chance.
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 |
| [2016-12-01 12:52:18] |
| skelcap - Posts: 139 |
|
Ok thanks, clearly it's not urgent. I wrote my own volume-based rollover algorithm for backtesting and was just using your values to check against. It's a minor inconvenience that my results disagree with your charts at times. I'm surprised though because I didn't think a volume-based rollover rule would need to change over time since it's not something set by an exchange. If you compare the contracts CBX15/CBZ15 side by side (attached) then it's pretty clear that the volume shifts between them around 10/12/15 and I would thing the algorithm should detect that. Having the roll dates out of sequence and skipping a contract on the chart just looks odd. |
 CBX15 - CBZ15.png / V - Attached On 2016-12-01 12:48:23 UTC - Size: 97.7 KB - 278 views Attachment Deleted. |
| [2016-12-02 23:01:23] |
| |
|
If you compare the contracts CBX15/CBZ15 side by side (attached) then it's pretty clear that the volume shifts between them around 10/12/15 and I would thing the algorithm should detect that.
Actually we now know what the reason for the problem is and we really cannot solve it. The date-based rollover was different in the past and there is a reliance on that in order to reliably determine a volume based rollover.
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: 2016-12-02 23:01:49
|
| [2016-12-03 02:58:21] |
| |
|
But we will think about a solution. It will take some time for us to figure out the best way to do this. The basic problem is the volume analysis is only going to be done within the particular month where the rollover occurs according to the current date-based rollover rules. Otherwise, the wrong rollover date can be determined. 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 |
| [2016-12-03 04:03:24] |
| skelcap - Posts: 139 |
|
Thanks I understand it is tricky and there is no one-size-fits-all solution. In many markets the shift in volume from one contract to the next seems well-defined, but in some markets like agriculture I have found there can multiple shifts which necessitates the use of some sort of date-based rule as well.
|
To post a message in this thread, you need to log in with your Sierra Chart account: