Request for a simple strategy to be coded for spread betting
-
This does what you describe but not profitably! Image shows FTSE100 daily. Even if you separate the long and short parts of the strategy it is not pretty.
defparam cumulateorders = false c1 = close > high[1] c2 = close < low[1] if c1 then buy 1 contract at market sell at low[1] - 15 stop endif if longonmarket then sell at low[1] - 15 stop endif if c2 then sellshort 1 contract at market exitshort at high[1] + 15 stop endif if shortonmarket then exitshort at high[1] + 15 stop endifHi, and thank you so much for taking a bash on the code for this. Can I just ask if this trails the stop loss with each new candle that is formed or if it fixes the stop at 15 below the low on entry? I only ask because when I run the backtest on the code you’ve given me it gets the entries just as I would do it manually, but the exits are different to where I’m taking them manually, so I feel like I’ve possibly left something out of my description or not conveyed it well. I have written on a screengrab of one trade example to highlight where my intended system differs from the code you’ve included. Follow the steps in the order I’ve numbered them. You can see where the PRT backtesting has entered and exited and where I’d need it to exit. Basically the SL is not moving along when a new candle is formed (it’s staying where it was initially placed). This is the major problem here from what I can see. Additionally, I notice this code deals in ‘contracts’ rather than ‘bet size’, would this still work for automating trading on a spread bet account within PRT? I hope this is helpful. Just want to reiterate how much I appreciate you taking the time to help with this. Regards MattIt shouldn’t be that difficult as long as you know how to trade manually and have the “insight”. Example below. The staircasing you see is a normal behavior. The climb vs the retrace (dropping back to a lower price level) is what you’ll try to cover with Trailing. So would you (manually) exit at say 13:24 ? perhaps. But would you at 13:36 after you experienced the part around 13:36 ? No ? then your Trailing should bear the distance to NOT exit at both the 13:24 and 13:36. Now, when that is just covered for, your Trailing will exit at ~14:36. This should not be found by optimizing, but by looking at the instrument’s behavior. Btw, I hope you both are serious and don’t turn out to be some kind of spammers (up to being the same person). We’ll see …druby thanked this post
- You must be logged in to reply to this topic.
New Reply
Author
Summary
This topic contains 15 replies,
has 7 voices, and was last updated by 
1 year, 4 months ago.
Topic Details
| Forum: | ProOrder: Automated Strategies & Backtesting |
| Language: | English |
| Started: | 09/22/2019 |
| Status: | Active |
| Attachments: | 5 files |
Loading...