Support Board
Support Board
Search Board
Control Panel
View My Posts / Threads
Direct MessagesDate/Time: Sun, 05 May 2024 05:00:14 +0000
Different backtesting results / order fills with original SC or TransactSC
View Count: 1590
| [2018-05-11 19:00:33] |
| impert - Posts: 6 |
|
I've created a trading strategy using spreadsheet autotrading with the original SC software I've copied all the studies and settings of this strategy to the SC version provided by Infinity/Transact The strategy uses a an attached target order for 9 pts or stop for 3 When backtesting on Transact's SC the algo often gets worse fills than when backtesting the exact same strategy on SC Here's 1 example, the top chart is original SC http://prntscr.com/jgp442 Looking at both spreadsheet's data at that candle, the data is the exact same General trade settings, General Settings, Symbol settings, Data/Trade Service settings are the exact same on both The issue is consistent and persists when changing the parameters of the stops or the studies it uses, the backtest on Transact always provides worse results Even when backtesting with Replay http://prntscr.com/jgpchc I've also tested connecting to Transact using original SC to download the chart data and test, but the results are still worse than using original SC I've tripled checked the parameters and any setting that im aware of, is there any setting that i might be missing? how can i best troubleshoot what is going on? Is there any difference between the 2 softwares in the way they calculate the backtests? |
| [2018-05-14 22:18:42] |
| |
|
Make sure Sierra Chart is set for a tick by tick data configuration: http://www.sierrachart.com/index.php?page=doc/TickbyTickDataConfiguration.php We really would not expect there to be much difference. Maybe some very small differences here and there in comparison to the data between services. We assume previously you were using the Sierra Chart Historical Data Service?: http://www.sierrachart.com/index.php?page=doc/SierraChartHistoricalData.php Maybe there are differences with the bid and ask prices stored with the trades. You can look at those through the Intraday Data Editor: http://www.sierrachart.com/index.php?page=doc/IntradayDataEditor.html 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: 2018-05-14 22:20:10
|
| [2018-05-17 16:26:36] |
| impert - Posts: 6 |
|
They are both connected to Transact so both are using the same data I checked the IntradayData and its the same tick by tick, i also redownloaded it Something strange that i noticed is the exact time of the fills In SC fills would always end in .500 or .501 for example 09:31:00:001 While in TransactAT they would always end in .000 or .001 This only happens though when using Bar based backtest, not when using replay based test but it also show different results between platforms |
| [2018-05-17 19:27:14] |
| |
|
Within a particular instance of Sierra Chart if you run the back test again and clear the prior Trading data, for the symbol and trade account, do you get the same result?
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 |
| [2018-05-18 22:22:28] |
| impert - Posts: 6 |
|
Yes, i just ran a test twice on SC and got the exact same results in both
|
| [2018-05-18 23:49:16] |
| |
|
Okay that is good. We are just still trying to think what the reason for the difference could be. So you are certain the underlying data in the charts are the same? Another variable that should not make a difference but we just want to make sure it matches, is the Global Settings >> General Settings >> Chart Update Interval assuming the chart is not overriding that and using its own setting. 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 |
| [2018-05-21 15:11:34] |
| impert - Posts: 6 |
|
Chart update interval is set to 500 in both
|
To post a message in this thread, you need to log in with your Sierra Chart account: