Discussing the strategy VECTORIAL DAX (M5)

Forums › ProRealTime English forum › ProOrder support › Discussing the strategy VECTORIAL DAX (M5)

Viewing 15 posts - 706 through 720 (of 1,261 total)
1 2 3 4 5 … 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 … 81 82 83 84 85
  • 03/19/2020 at 12:17 PM #122570
    saturnin

    ok. thanks you but backtest 5 pertes=1000 euros demo 1 perte=55euros

    03/19/2020 at 3:05 PM #122585
    nonetheless
    fifi743 wrote:

    put the mouse on it and you have the operation number

    The values for AC vary between 0 and 16. Should AC be >0 ?

    03/19/2020 at 4:35 PM #122596
    fifi743

    NO
    It gives you where the problem is.

    03/19/2020 at 4:58 PM #122602
    nonetheless

    I couldn’t see anything unusual on the graph. Maybe it was just a blip? It has been working fine since that one rejection, taking trades.

    Actually, for now it’s outperforming both the DAX v8, and the NAS v6 but much too early to judge. Thanks for your help!

    03/24/2020 at 4:43 PM #123168
    Nacho

    Can you upload the .itf file with which you get that backtest? I have tried with the different V6 and none gives me those results and there is one that gives an error.
    Thank you.

    04/14/2020 at 10:15 PM #126170
    nonetheless

    The DAX 5m version I was running went a bit flat from Nov so I refreshed the numbers and it’s looking better than ever. Image on the left is (I think) the last version posted with mods by Paul and Fifi. I didn’t touch any of their work, just re-optimized the vectorial settings @ 60k. So the image on the right is effectively a 30-70 walk back. Mostly improved, especially over the last 6 months.

     

    2 users thanked author for this post.
    avatar Fred23, avatar keewee
    04/14/2020 at 10:20 PM #126174
    nonetheless

    So once I did that, I figured I’d better do the NAS 5M as well. Made a few small changes and re-optimized over the full 200k. Robustness looks acceptable, but could probably be further improved.

    Thanks again to Balmora for this exceptional algo, and to everyone else who contributed.

    2 users thanked author for this post.
    avatar Fred23, avatar keewee
    04/15/2020 at 12:26 AM #126190
    Francesco
    nonetheless wrote:

    The DAX 5m version I was running went a bit flat from Nov so I refreshed the numbers and it’s looking better than ever. Image on the left is (I think) the last version posted with mods by Paul and Fifi. I didn’t touch any of their work, just re-optimized the vectorial settings @ 60k. So the image on the right is effectively a 30-70 walk back. Mostly improved, especially over the last 6 months.

    Robustness went worse with this version

    04/15/2020 at 6:44 AM #126196
    MAKSIDE

    another version that i have tested about DAx vectorial.  an MTF version of it.

    lower gain but less ddmax and more consecutive gains

    i had noticed during the big period of volatility  that parameters had a positive impact between february and april.

    it might be interesting to store the value of these parameters in several variables, a variable per digit, to count the parameter used during the profit or not, and as soon as there is a loss or a gain depending of strategyprofit, to apply a weight of the parameter, then, 2 possibilities,  average of the number stored following the best weights or average to take the parameter with the best weight..  a possible evolution of machine learning

     

    04/15/2020 at 8:30 AM #126203
    nonetheless

    @Francesco here’s the results i got. I didn’t bother posting it because I’m not at all confident that I got all the tradeons in the right places — there are an unholy number of entry points.

    The way I see it though, you’re looking at a 140k OOS back test with a better curve, better profit, more trades, much lower DD and a correction to the recent months when it has frankly been quite dead … where would you rather put your money? The robustness score would have to be utterly disastrous to change my mind.

     

    04/15/2020 at 8:44 AM #126205
    nonetheless

    @MAKSIDE what TF did you add and what did you put there?

    I also tried experimenting with MTF, here’s one that might be interesting — doubling the vectorial component at the 15m TF. Very few trades, as you would expect, but 100% wins. I’m sure it can be improved if someone wants to take it on…

    2 users thanked author for this post.
    avatar Fred23, avatar MAKSIDE
    04/15/2020 at 10:20 AM #126222
    MAKSIDE

    @nonetheless

    thx for sharing ! nice. if 100% wins, we can play with the positionsize.. 😉

    about my version, pls see below..  cbuy and sell added to the current conditions

     

    04/15/2020 at 10:52 AM #126233
    Francesco

    I got those results..i’m pretty sure i put tradeon on every single entry condition

    04/15/2020 at 11:01 AM #126239
    MAKSIDE
    Francesco wrote:

    I got those results..i’m pretty sure i put tradeon on every single entry condition

    is it the nonetheless’s version ?

    04/15/2020 at 11:03 AM #126240
    Nicolas

    Just a test of upload of a picture.

  • Author
    Posts
Viewing 15 posts - 706 through 720 (of 1,261 total)
1 2 3 4 5 … 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 … 81 82 83 84 85

Find exclusive trading pro-tools on ProRealCode Marketplace

Similar topics:
adding input from custom indicator into strategy
array/bubble sort in strategy doesn’t work, but works as an indicator
auto restart automated strategy
bb trading one long one short on dax 9-17,30
bollinger+fractal strategy idea
can you add compounding to a strategy?
central bank dealers range – reversal strategy help please!
combining short/long strategies into 1 strategy issue
conversion of a trend following strategy from the multicharts trading software
conversion of pinescript strategy (tradingview)

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