Back testing strategies

Forums › ProRealTime English forum › ProOrder support › Back testing strategies

Viewing 15 posts - 1 through 15 (of 104 total)
1 2 3 4 5 6 7
  • 06/21/2018 at 5:31 AM #73844
    ChrisNYE

    Hi guys,

    I know that developing strategies can be done and done with high % returns.

    For the life of me I cannot get anywhere near the returns I see people get on this forum. So clearly i am missing something.

    I am seeking to be pointed in the right direction. I currently use a 200 ema as a trigger with a 2 ema cross. What else can I look into using as an indicator to assist my trading. I really need help this is starting to really get to me. I am not asking for your code, just to be pointed in the right direction.

    06/21/2018 at 7:35 AM #73851
    Vonasi

    First of all you need to know what you are trying to achieve. Without knowing that is like heading off on a long trip without a map.

    Are you trying to build a mean reverting, trend following, breakout or other type of strategy.

    Is it intraday, end of day or even end of week or month?

    Short time on the market or longer hold?

    Based on price action or oscillating indicators or a bit of each?

    After a while you will find what suits you – but only after a few thousand failed attempts. If you have not yet got 1000 strategies that you already created that do not work then you are nowhere near ready to go live with real money!

    If I ever run out of ideas to test then I just hit Google and search out other peoples blogs and webpages and see what they are trying out. Coding someone else’s idea is a good challenge and very educational and you have the benefit that they have already tested it for you so you know whether it is worth continuing with from their report.

    1 user thanked author for this post.
    avatar GraHal
    06/21/2018 at 7:39 AM #73853
    Vonasi

    You might want to start here for some ideas of where to look for trading ideas:

    https://www.prorealcode.com/topic/interesting-web-pages-or-blogs/

    06/21/2018 at 8:15 AM #73859
    GraHal
    Vonasi wrote:

    If you have not yet got 1000 strategies that you already created that do not work then you are nowhere near ready to go live with real money!

    I think above is so true and a useful mantra for newbie Auto-Traders to recite and remember.

    I’ve always learnt more in life by my mistakes … I always ask myself what went wrong and  try and remember to never repeat the same error! 🙂

    06/21/2018 at 8:47 AM #73867
    jebus89

    Yea u just gotta keep creating, but just using 2-3 moving averages is basicly suicide in this game.

    At least i have never seen a good strategy using only a couple moving averages.

     

    Keep testing every single indicator in every single possible way u can think of.

    Add an RSI and add a moving average on it and see what u can use that for. Maybe as a good filter? Maybe not usefull at all?

    Just read about indicators, study what they do and how to interperet them. Watch videoes on how other ppl are using it. Look at algos in this forum and how they work. keep at it.

    it takes time and dedication and everyone was once at the same place u are now. But if u keep giving it 100’s of hours and as mentioned, try to create 1000 strategies and yes 995 of them will probably look like shit or go nowhere, but those last 5 might be ur bread and butter for the next 5 years…

    06/21/2018 at 9:27 AM #73870
    ChrisNYE

    Thanks guys looks like a long journey ahead.

    06/22/2018 at 3:52 AM #73978
    ChrisNYE

    As I want to be a trend trader I need to develop a strategy that accomodates trend following entries. So I need to develop a strategy that can place trades when the instrument is trending and if the market is range bound to sit on the side lines.

    Is it possible to develop a system that has 2 strategies. One for if a trend is present and one for if the market is ranging?

    Chris

    06/22/2018 at 6:47 AM #73979
    GraHal

    Why complicate it? It’s hard enough getting success with a one type strategy?  🙂

    Why not have 2 separate strategies in 2 separate Auto-Systems?

    06/22/2018 at 7:38 AM #73982
    jebus89

    Yes u can make “multiple strategies” inside 1 strategy but its usually much easier to just make 1 trending and 1 for choppy markets.

    If ur a “trend follower” u need to filter out when its a TREND market and when its a CHOPPY market.

    U need to enter as early in the trend as u can and exit with profit before price falls too much. Just think about that as u stare at the chart.

    06/22/2018 at 8:03 AM #73989
    ChrisNYE

    Okay, so my next question is how do you make consistent profits with any strategy if we know both trends and range bound markets never last?

    I have developed strategies that make large profits when trading then absolutely nothing for months as there is no real trend, but on the flip side I have no losses either – trades just dry up. Take for example the EUR/USD since early April a great short. BUT take the same fx pair and between Jan to April and it was range bound. This is why I am going crazy – trying to develop 1 strategy in back test that has consistent profits year after year with the same strategy I believe is near impossible UNLESS it is in one huge obvious trend OR like you have both pointed out you have 2 different strategies one for trends and one for ranges.  But this will not show up as an amalgamated P&L in auto back test.

    Pls tell me I am moving forward with this analysis 🙂

    Christian.

    06/22/2018 at 8:14 AM #73992
    GraHal
    ChrisNYE wrote:

    But this will not show up as an amalgamated P&L in auto back test.

    But if want to monitor two separate Systems as an integrated pair then just add INT1 & INT2 (or something similar) in the titles of Systems and add the two P&L across the two of them??

    If you get it right then you should see action (broadly) on Trend but not Range System and vice versa!?

    Yes you are moving forward, but tell us you are not trading Live yet!? 🙂

    06/22/2018 at 8:25 AM #73995
    Vonasi

    Start with one strategy that might only place trades during strong trends or is good at breakouts into a trend and then when you have mastered that develop a mean reversal strategy for the times when the same market is range bound. Then get those ideas to also work on other markets of the same type. Then get them to work on commodities, forex and indices. Then you have a whole portfolio of strategies and if one is not trading or losing then the others will hopefully be compensating for it. Your overall equity curve will be much smoother (hopefully if at least the majority of strategies work well!) Forward test for months until you are satisfied and confident that you have a good suite of strategies across a broad range of instruments.

    It is called diversification – or not having all your eggs in one basket.

    06/22/2018 at 8:31 AM #73996
    ChrisNYE

    No back testing only = Put the funds back on the mortgage.

    Andy my goal is to develop trend and mean reversion strategies for fx majors and indecies.

    I see your point if I add up all systems overall I must be in profit as a few may be a loss for a period but others may be in a larger profit. So net profit is the aim

    06/22/2018 at 8:32 AM #73997
    ChrisNYE

    So when I am told to make consistent profits year on year they mean as a whole and not individual strategies?

    06/22/2018 at 8:52 AM #74004
    Vonasi
    ChrisNYE wrote:

    So when I am told to make consistent profits year on year they mean as a whole and not individual strategies?

    Almost certainly…. and they have been trading for years and developing their experience at what works and what doesn’t and what to look at and what not to bother with and how to code it efficiently. Once you reach this point the working strategies become easier to find and spot and test and the rubbish can be discarded quicker.

    As has been said many times on here before – trading is one of the hardest ways to make easy money. You have to enjoy doing it and have a lot of time to dedicate to it. I view it as a fun hobby with the added bonus that I may end up with more money than I started with. I am not an equity fund manager trying to beat the market. If I want to earn 7% then I will just put the money in P2P lending and hope the world does not go into a major recession!

    Your idea of paying off your debt is a far better idea than using the money for trading until you are confident that you are actually quite good at it! Debt should always be got rid of first before anything else is done unless you can find a guaranteed income % greater than your mortgage % IMHO.

  • Author
    Posts
Viewing 15 posts - 1 through 15 (of 104 total)
1 2 3 4 5 6 7

Find exclusive trading pro-tools on ProRealCode Marketplace

Similar topics:
back testing error as you chack longer periods
extend the time period for back testing
rookie question — back testing issues
adding entry filters to your strategies
combined mean strategies
combining short/long strategies into 1 strategy issue
create strategies codes with chatgpt for prorealtime
division by zero parameter error on 3 different strategies
help with backtest that only goes back a few days
how can trial strategies purchased from marketplace be deleted

Create your free account now and post your request to benefit from the help of the community
Register or Login