Don't risk your money on myths

Discussion in 'Technical Analysis' started by Murray Ruggiero, Aug 10, 2011.

Do you use backtesting ?

  1. I don't know how to do it.

    7 vote(s)
    11.1%
  2. I think backtesting is a fraud

    14 vote(s)
    22.2%
  3. I use it for simple systems but not my chart analysis

    7 vote(s)
    11.1%
  4. I would not trade anything without testing it

    35 vote(s)
    55.6%
  1. Murray Ruggiero

    Murray Ruggiero ET Sponsor

    Most trading ideas in books,video and expensive classes simply do not work. They give you rules , indicators but never show you results on historical data.

    Backtesting: Running a set of rules on historical data is called _backtesting.

    Most indicator products, only give you indicators, they don't let you build system so you can test them because they don't work.

    The market is a zero sum game except for slippage and commission.
    Most indicators break even or worst , that why they make sure you can’t backtest them.

    We will explain how to use technical analysis and both backtesting and walk forward testing to give you the best chance to be a successful trader.
     
  2. Samsara

    Samsara

    Hi Murray. Glad you're offering a demonstration of TS in this thread; looking forward to it.

    I have a tangential question: does TraderStudio allow simultaneous backtesting of multiple instruments in one pass? For instance, signals generated from logic on a basket of commodities and ETFs.

    One thing eSignal could not do for me was allow querying of more than 7 symbols. I'm wondering how to approach that task without building a custom SQL setup.

    Much appreciated.
     
  3. Murray Ruggiero

    Murray Ruggiero ET Sponsor

    Are we talking about intra-day testing or end of day analysis
     
  4. Samsara

    Samsara

    Both.
     
  5. Murray Ruggiero

    Murray Ruggiero ET Sponsor

    For end of day data I use both Pinnacle for Futures data and CSI for stocks. I also have CSI world futures for many international markets.

    In terms of E-Signal I know there is a utility to export the data into CSV files, I think it's QLink but I am not sure.

    You can import this data into TradersStudio.
     
  6. Murray, does TS have the capability of building accurate Constant Vollume bars and if yes, can they span over more than one day?
     
  7. Murray Ruggiero

    Murray Ruggiero ET Sponsor

    No, I don't build constant volume bars. If I have demand for it I might add it in the future.
     
  8. Thanks for the response.
    In the beginning doctors didn't think much of penicillin either.
     
  9. Murray Ruggiero

    Murray Ruggiero ET Sponsor

  10. I believe it's been done but I've never seen the results. The biggest problem is coming up with objective rules. How do you identify the swing highs/lows to determine Fib levels? Do you always exit at other Fib levels or do you use stop losses, etc?

    I remember Prechter saying he studied Fib levels exhaustively and didn't find them profitable. Of course, his own Elliot work has been quite subjective/biased since he became a permabear over 20 years ago.

    Often, "Fib traders" say it only works when there's a confluence of Fib levels, other S/R, candlesticks, round numbers, sunspots, winning Lotto ticket numbers and who knows what else. Of course, it's pretty much impossible to test such "confluences." I've asked some of them to post live entries/exits for a statistically-significant number of trades. Not surprisingly, none have taken up that challenge...
     
    #10     Aug 12, 2011