Ok so I only started using PRT (via IG) in the past few days. Been testing systems on Australia 200 cash, but looking to build systems that trade on a few other indices too. I’ve read as many posts as I can find about strategy creation, and how to approach back testing here.
Wanted to ask a few questions before I go deeper.
(1) Timeframes on a 100,000 bar limit
What is the shortest time frame that you find you are able to produce robust systems in, with the 100,000 bars limit in IG? Is there a threshold? i.e. On 1 minute bars, I can only get around 3 months of data, which basically gives me no idea as to how it would trade before this Covid-19 crisis. With 5 minute bars, I get to go back abit over 1 year… but it’s still 100,000 bars. With a couple of filters and entry rules, I only end up with around 100-200 trades, sometimes less.
(I know this will vary a lot, and depend on personal preference/trading style, but just want to hear about other experiences)
Which leads to my next question:
(2) Backtests on index CFDs vs underlying index futures
Has anyone gone down the route of purchasing futures data to backtest (perhaps on some other systems such as MT4 / Amibroker), and compared results to backtested IG index CFD products in ProRealTime? Asking because I want to know if the results are comparable. My understanding is that they are different products, but IG makes the price movements as close to the futures as possible – but to what extent in the real world, I have no idea. If this has been tried, and results are completely unreliable, then I will steer clear of spending time and money going down this route.
(3) Sub-minute time frames
Just for my own curiosity – does anyone use systems that trade on say, 5 second or 10 second time frames? I’ve been able to make a few that seem to do ok, but have a feeling they are very curvefit, despite just using a couple of entry triggers and 1 or 2 trend filters. Time will tell!
.
