Mr. Anand HS Forex Devisen
We asked Mr. Anand for a high-level description of one of his systems. Here provided an outline that summarizes one of his most profitable ones:
- At the end of the trading day, calculate several values with different time-frames for the Stochastic, and RSI. Calculate Z-Score and CCI General also.
o If both (RSI and Stochastics) are oversold then a long signal is indicated.
o If both (RSI and Stochastics) are overbought then a short signal is indicated. Z-Score must be negative (preferably –2) for a long signal and positive (preferably +2) for a short signal.
o CCI General must be similarly negative for a long signal and positive for a short signal.
o The above mentioned indicators may be appropriate as inputs for a Neural Network.
o Ensure that the current trend (3Bar/10Bar EMA) is in the opposite direction. Remember we are going against the trend because we are looking for Trend changes based on the “Contrarian” theory.
o The above steps are guaranteed to filter a top or a bottom and tell us what signal to look for (buy or sell) and when to look for that signal. This solves 50% of the problem of timing the market tops and bottoms.
- Once he determines what signal to look for (buy or sell) and when to look for a signal, he then searches for those signals using an Entry/Exit system based on a proprietary algorithm which has been in R&D for over 45 years. This solves the other 50% of the problem. Actually, you can use any Entry/Exit system to detect signals, including Traditional canned systems and Advanced Neural Network based systems with the appropriate inputs.
- Verify the signal detected with at least one other signal in a different time-frame (3 bar with 4 bar with 5 bar etc.). Verification is more difficult than Detection.
- Interpret the detected and verified signal as to whether it is a short-term or an intermediate-term or a long-term signal. Ignore short-term signals. Short-term profits lead to long-term ruin. Interpretation is more difficult than Verification.
- Once the signal is verified and interpreted, enter the trade at Market On Open (previous session’s Market On Close) with a 1-point protective tight stop (for the current session only) for intermediate-term trades and no stops for long-term trades. For long-term trades preferably combine an inexpensive put or a call option with the straight underlying long or short instrument, thereby simulating an expensive call or a put option respectively.
- Use Gann’s Rule of Eights stopping methodology for Intermediate and long-term trend changes for the next session after entry.
- Try to put a Target profit limit also, by using the Medium Term (32 Bar) Gann swing points and determining the limit by an appropriate method.
These three steps (Detection, Verification and Interpretation) combined with efficient money-management techniques are the key ingredients for what he refers to as his “ultimate trading machine”. So far he is not using any Neural Network Models because he only has three years of historical Forex data and would like to have at least five years of data in order to adequately train the networks. Based on the success of his rule-based systems however, a similar neural network approach using the appropriate inputs may likely produce similar spectacular results.
Kann mir jemand helfen und sagen wie ich das in Investox umsetze ?
