Is there anyone out there who can say they make money with automated trading?

Forums › ProRealTime English forum › ProOrder support › Is there anyone out there who can say they make money with automated trading?

  • This topic has 51 replies, 21 voices, and was last updated 6 years ago by avatarAbz.
Viewing 15 posts - 16 through 30 (of 52 total)
1 2 3 4
  • 09/30/2017 at 3:23 PM #47877
    Francesco78

    this is an example of my equity lines. In real. All with prt.

    1 user thanked author for this post.
    avatar Andyswede
    09/30/2017 at 5:04 PM #47887
    Despair

    @Francesco: Which timeframe you run this strategy on? 45 trades per day is A LOT!

    09/30/2017 at 5:54 PM #47890
    Francesco78

    @despair : it is a portfolio of 75 strategies, ranging from 5 minutes to 1hr. 70% of them are on 1 hour TF

    2 users thanked author for this post.
    avatar reb, avatar Despair
    09/30/2017 at 6:27 PM #47892
    Despair

    OK, that is for your whole portfolio. I see. I thought it was a single strategy and was really impressed by the number of trades hahaha.

    09/30/2017 at 9:55 PM #47899
    juanj

    Very interesting, I thought that by running 6 strategies live it was allot. Haha clearly you have some decent capital to play around with. How many different instruments are you trading across those 75 strategies and does it include shares?

    1 user thanked author for this post.
    avatar Francesco78
    10/01/2017 at 4:24 PM #47935
    cpgraham

    Yes. I agree with the advice above. It takes a lot of dedicated time and effort.

    10/03/2017 at 5:35 PM #48138
    ozz87

    Which strategies have you got good results with? 🙂 I’m running pathfinder swing but got to low capital for it (so far lost money).

    10/03/2017 at 11:05 PM #48147
    Despair

    I think like Francesco also already pointed out that diversification is the true holy grail. It is more a portfolio of mediocre performing strategies running simultaneously than to find the one strategy that will make you rich.

    2 users thanked author for this post.
    avatar manel, avatar ozz87
    10/04/2017 at 6:14 AM #48152
    ozz87

    @Despair: Yes, true. I saw his screenshot 😉 Testing some new ones on my demo account. Looking for some that requires low capital, otherwise it would be hard to run more than just a few. Also trying normal pathfinders but changed to max 1 contract/trade. The problem you pointed out with using old data with Pathfinder Swing doesn’t look as bad with them.

    10/04/2017 at 11:21 AM #48178
    victormork

    @ozz87 I think what a lot of people forget before they start to trade a system like Pathfinder is that;

    1. You risk a lot of money on each trade just because you want to be more right than wrong. It feels good to be right but trading are about making the max amount of money out of each dollar/euro/pound or whatever else you risk.. Just because it had a 90% win rate in the past doesn’t mean that it will be the same in the future. The win rate might even be less than 50%. What will happen to your account when you have a 50% win rate and risk 3 times what you can win on each trade?
    2. Also a system like Pathfinder can stay under water for a very long time without being broken.. Are you prepared lose money for a loooooong time?
    3. A lot of backtests I’ve seen for different version of the Pathfinder have very few trades. So what is the probability that all bad trades were filtered out in the past but might get taken in the future?
    4. @Despair is coming with a very good point that it’s not ONE system that will make you rich. Rather focus on diversification and position size mgmt. 10 average systems can together beat one good system. This can also smoothen your equity curve.

    I’m not trying to take a shit on pathfinder. It’s just an example. Mean reversion strategies tend to have the same issues as I pointed out on pathfinder.  I just want to explain why you should not make the same misstake that I, and many more people, have done in the beginning. There are great tools out there for system evaluation. USE THEM BEFORE YOU RISK YOUR MONEY! ( https://www.prorealcode.com/topic/system-quality-number-evaluation-of-performance-of-algorithm/ )

     

    2 users thanked author for this post.
    avatar Ste, avatar ozz87
    10/04/2017 at 11:55 AM #48182
    ozz87

    @victormork: Yes, pathfinder seems to be pretty good, but I don’t have the capital to run 10 of them. That’s why I’m looking for systems with less money required to get more diversification. Trying some different ones on my demo acc now, fractal and the “1 minute tactic” for example.  Actually I’m looking for more systems with abit less capital required. Anyway thanks for your advice 🙂

    10/04/2017 at 12:30 PM #48185
    Despair

    Find a system that produces a good WFA or at least one OOS run (70/30).

    Then run a monte carlo simulation with the OOS results you got to determine the equity requirement to trade the strategy safely (I always want the risk of ruin to be less then 10% within the first year).

    Then run this system a few months on demo to see if it does there the same as it did in backtest. If after a few months the demo results are consistent with the backtest you are good to go to start live trading.

    3 users thanked author for this post.
    avatar Ste, avatar AsgerK, avatar ozz87
    10/04/2017 at 12:54 PM #48190
    victormork

    Spot on @despair! Aslo try a few different markets for the same system. If a system works on DAX it should probably work on DOW, Nastaq etc. If you have  system for EURUSD then what about GBPUSD, AUDUSD etc?  It doesn’t take a lot of effort to validate.

    1 user thanked author for this post.
    avatar ozz87
    10/04/2017 at 3:07 PM #48202
    ozz87

    I am trying some OOS testing now, just learned how to do it. But the results vary alot. For example on pathfinder for ftse I got WF for my tests, in%: 245, 113, 140, 16, 49 . However I’m not sure how the variables should be filled in for the tests (used the ones in the PRT guide).

    Sorry for offtopic…

    10/05/2017 at 10:00 PM #48344
    Vonasi

    Thank you for sharing your equity curve/s Francesco78. It created a bit of a ‘Eureka’ moment for me. The search for the massive money making holy grail is over and all those strategies that looked like they would make a little money now and then will now be given a new lease of life. Just need to find some more capital!

    On the subject of diversification – what are peoples thoughts on running one strategy with different optimization results multiple times. For example you optimize a strategy and by changing just one number in the strategy you might get say more winners with one number or a higher overall gain with another number or a higher average gain per trade with another number. Why not run 3 strategies and average out the returns?

  • Author
    Posts
Viewing 15 posts - 16 through 30 (of 52 total)
1 2 3 4

Find exclusive trading pro-tools on ProRealCode Marketplace

Similar topics:
is there a pr api that can send offline alerts (sms-email
can’t find automated trading
is there a “search” button here ?
is there a easier way to hide the code on algos?
is there any point devising strategies and backtesting?
is there any way to convert this indicator to prorealtime?
is there such a strategy existing in autotrading?
combining manual trading and auto trading
difference in outcome backtest and automated trading
gain today? why when there is no position open?

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