trailing or fixed stops vs variable (ATR based) stops

Forums ProRealTime English forum ProOrder support trailing or fixed stops vs variable (ATR based) stops

This topic contains 2 replies, has 3 voices, and was last updated by avatar AlexF 8 minutes ago.

View all attachments
Viewing 3 posts - 1 through 3 (of 3 total)
  • #36465

    Hi all,

    I am wondering wheather anyone has considered this probelm, I wanted to do some reaserch about it and I asks in case someone has thought about it.

    In considering trailing stops vs ATR multiplied stops in backtesting,  have you ever recognized some sort of correlation between the results of a backtest using trailing or ATR multiplied stops and general market condition? (lateral, trendy etc..)

    Many thanks

    Francesco

     

    #36485

    Great question, it’s something I have given a lot of thought to.   Basically what we are looking for is a means to make parameters dynamic so that they are representative of the conditions they are running in.

    I have tried (and continue to try) many methods and have applied them to stops, limits, my manual trailing stop levels, entry conditions etc.  I always split the long and short trades with separate stops etc. for each.  A couple of examples of things I have tried:

    Market sentiment,  Bullish or Bearish based on some simple criteria, using hard coded stop values etc. but different values (as discovered through backtesting) based on current market sentiment.  Could be just more curve fitting but logically it makes sense (to me at least) that e.g a limit should be set to be less for a long trade in a bearish market.

    Volatility.  In some strategies I took the mean of the last 3 day ranges divided it by a constant to create a multiplier and applied that to stops etc.  Works ok (in backtesting at least) on index futures.

    Market condition.  Love to use this but I don’t know how to determine the various markets conditions, probably just me but even identifying a trend seems problematic.

    Look forward to hearing what other methods are being used.

    1 user thanked author for this post.
    #78598

    Volatility.  In some strategies I took the mean of the last 3 day ranges divided it by a constant to create a multiplier and applied that to stops etc.  Works ok (in backtesting at least) on index futures.

    Hi, would you share this code?

    Thanks

Viewing 3 posts - 1 through 3 (of 3 total)

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