Chabah on Automated Trading

Discussion in 'Automated Trading' started by Chabah, Oct 2, 2006.

  1. I've read Van Tharp several times myself and I agree with the 1% rule. But Van Tharp was also a big proponent of high R-multiple trades. So according to him you should be looking for at least 2R or 3R trades, regardless of what % of capital R represents.

    It looks like you're looking at 1R trades, unless I missed something.
     
    #21     Oct 3, 2006
  2. Chabah

    Chabah

    I'll be using several strategies, but in general I am looking for large R-multiple trades, not 1R.

    Some of the glossary definitions imply ways to defend capital, but those are concepts, not strategies. I will expand on this soon - I already have it written up but would like to review before posting.

    Thanks for the feedback!

    Chabah
     
    #22     Oct 3, 2006
  3. Ami-Chai

    Ami-Chai

    Hi Chabah,

    I would be interested in learning more about your trading strategy, i am interested in finding automated systems to invest in.
    Contact me if you would like to discuss this further.
    Ami
     
    #23     Oct 3, 2006
  4. Chabah

    Chabah

    Research

    Most research will be automated by the individual trading strategies. Manual research will be based on services such a The Street (Cramer) and Real Money, forum/blog watch lists, and possibly other subscription services to identify candidate securities. The focus is on automated research.

    Each strategy will generate signals that fall into an output category. Some sample Research Outputs include:

    Long Setups
    Short Setups
    Trend Setups
    Pair Setups
    Pattern Setups

    Long Setups are based on an anticipated upside breakout. If the product does not increase in value, the setup is invalidated and should be exited ASAP.

    Short Setups are based on an anticipated downside breakout. If the product does not decrease in value, the setup is invalidated and should be exited ASAP.

    Trend Setups, or momentum setups, are based on anticipated strong movement either up or down. These products may be entered once there is a discernible trend and may be reversed (not just exited) when the trend reverses.

    Pair setups are based on price correlation between two stocks. These may be extended to group setups as well.

    Additional setups will be added as I learn about them.

    __________________________________

    I understand that this is basic stuff, but I want the thread to be accessible to both beginners (like myself) and experts (most of you) alike.

    Chabah
     
    #24     Oct 3, 2006
  5. fletch2

    fletch2

    I have a similar background as you.

    My only comment so far is it seems you've done an awful lot so far without doing the only thing that really matters - finding an edge. That's the whole game. All the details you're discussing are just window dressing on your edge, if you can find one.

    Fletch
     
    #25     Oct 3, 2006
  6. Chabah

    Chabah

    Thanks Fletch, that is a good point. I have worked with software systems for years, and I find that spending time up front to make sure the architecture is flexible and scalable is well worth it. I think it is more than window dressing.

    I will start with a system that handles all the research, signals and executions flawlessly and is ready to handle multiple strategies. The starting strategies are straight out of Professional Stock Trading by Mark Conway and will have little or no edge.

    My edge(s) will come as I start to study the results and alter the strategies to my specifications. This is knowledge I don't have yet. I also have a trading partner (separate account) with whom I can collaborate for ideas.

    Your assessment is correct but not unaccounted for. If I never get an edge, then I'll lose my money and stop trading. For now I need to complete the foundation and architecture.

    Thanks!

    Chabah
     
    #26     Oct 3, 2006
  7. Chabah

    Chabah

    Entry

    Entry is the act of opening a position based on an input signal from Research. Not all signals result in an order since the signal must pass through a position sizing algorithm and validation checklist.

    The required inputs from Research are the product symbol, Purchase Price and Initial Stop. Other variables that may be available at the time of entry are the Cost Stop, Risk Stop, Pure Profit Stop, and Closing Stop. See Glossary for term definitions.

    Position Sizing

    To determine the Position Size it is necessary to consider several variables, including Active Capital, Cash Available, Priority, Purchase Price, Initial Stop, Target Risk Percent, Target Risk Amount, and Actual Risk Amount. Some of these are inputs and some are calculated values or results. See Glossary for term definitions.

    The Target Risk Percent is calculated by multiplying the Priority by .1. For example, a Priority 3 stock has a Target Risk Percent of .3%. The Target Risk Amount is calculated by multiplying Active Capital by Target Risk Percent. The Target Risk Amount is then divided by |Purchase Price - Initial Stop| to yield the Position Size.

    For example, for a $50k Active Capital account with a Target Risk Percent of 1%, the Target Risk Amount is $500. If the stock is $10 and the Initial Stop is $9.50, divide .50 into $500 for 1,000 shares.

    Checklist

    The Checklist validates the Position Size to ensure that it meets system rules. It also may check market conditions and nullify or amplify the trade. Checklist items include (but are not limited to):

    Max Position Percent: If Position Price exceeds rule for maximum percent of Active Capital, reduce Position Size so that Position Price meets rule specifications.

    Cash Available: If Position Price exceeds Cash Available
    1. Raise cash by closing lower priority positions or
    2. Reduce Position Size so that Position Price = Cash Available

    Existing Position: If there is an existing position in the same product in the opposite direction, then the existing position should be closed without affecting the new Position Size. If the existing position is in the same direction, then the new Position Size should be reduced by existing position size. All shares are then considered part of the new position (existing orders for the old position are cancelled).

    Size Check: If Position Size or Position Price is less than a minimum value, abort trade.

    Account Losses: If daily account losses exceed limits set in the rules, abort trade.

    If the Position Size is changed during the checklist the entire checklist is repeated. Once the Position Size is determined, the order can be placed. The Initial Stop order is placed immediately.

    _________________________________

    I am aware that this is a pretty basic position sizing algorithm - this will be updated and/or expanded as I am exposed to others. If you have any good suggestions, either specific methods or just books or websites to read, please leave a comment.

    Chabah
     
    #27     Oct 3, 2006
  8. Ami-Chai

    Ami-Chai

    i must agree finding an edge is the most important aspect of trading.
    once you have that you are on you way...
     
    #28     Oct 4, 2006
  9. Your edge is in the sum-of-the-parts of the whole system.

    Your edge can/will be found as you continue to do the work involved with creating it.

    There are only "X" number of strategies available to traders in the marketplace, the number is finite (though longer than I know, but still finite).

    Through your thorough and exacting approach to trading from the very beginning I am confident that you will find at least one of those edges, if not more.

    Regards,

    Jimmy Jam
     
    #29     Oct 4, 2006
  10. Chabah

    Chabah

    Unwinding

    Unwinding is the process of closing an open position, including the intermediary goals of capital protection, cost protection, risk protection, earnings protection and profit taking.

    There are multiple unwinding strategies that are separate from the research strategies (although some may intentionally fit together). The strategies share common order types, including the Initial Stop, Cost Stop, Risk Stop, Pure Profit Stop, Closing Stop, and Time Stop.

    The Initial Stop serves the goal of capital protection. The stop is placed at the time of order entry and is never pulled or moved away from the price, although it is reduced or cancelled when other stop loss orders are placed.

    The Cost Stop serves the goal of cost protection, including the commissions for the opening order, the Cost Stop execution, and at least one additional execution (more if multiple unwinding steps are expected). The Cost Stop is expressed as a price for a certain number of shares, e.g. $20.05 for 100 shares or $20.10 for 50 shares.

    Example: Trade fee is $10, Position Size is 100, and Fill Price is $20. In this case the Cost Stop needs to cover $30. Therefore the Cost Stop for 30 shares is $21. The Cost Stop for 60 shares is $20.50.

    The Risk Stop serves the goal of risk protection. Risk Stop calculations are directly proportional to the Target Risk Amount, or R. Risk Stops are expressed as a price for a certain number of shares.

    Example: Purchase Price (and Fill Price) $10, Initial Stop $9.50, Position Size 100 shares. The Target Risk Amount (and Actual Risk Amount) is $50, or 1R. The 1R Risk Stop for 100 shares is $10.50. The 2R Risk Stop for 100 shares is $11. The 1R Risk Stop for 50 shares is also $11.

    Risk Stops are generally incremented along with the price, and may be implemented as a trailing stop or set of trailing stops. The idea is to first eliminate risk and then to lock in small profits. Note that by “eliminate risk” we are only referring to executions of Stop Loss orders – this provides no protections against gapping. Risk Stops may include commissions and predictable (average) slippage as applicable.

    A Pure Profit Stop serves the goal of earnings protection. The intention is to close a portion of a position and in doing so recover the original cost of the entire position (Position Price) including all commissions. The fewer shares sold, the better, as the remaining position will earn all profits.

    A Pure Profit Stop may often be executed as a Market or Limit order, particularly near the market close. The risk of gapping down and losing a guaranteed profit can be avoided by executing a Pure Profit Stop, as the gap risk is then only applied to the profit (not the entire position).

    It is especially desirable to take profits when other market opportunities are available – that is, when other orders are not being executed due to a lack of Cash Available. A Pure Profit Stop frees capital for other trades while allowing the existing trade to earn further profits.

    The Closing Stop serves the goal of profit taking, and is the final sell order for a position. Since Initial, Cost, and Risk stops often close a position, the main time a specific Closing Stop is required is after a Pure Profit Stop. The Closing Stop may be implemented as a trailing stop. Note that according to the trading rules the Closing Stop should never be adjusted away from the price.

    A Time Stop is a conceptual order used to control existing stop loss orders. Using a Time Stop in a strategy allows positions to be closed without hitting the stop loss order(s), for example if there are only slight gains or losses. A Time Stop can ratchet existing stop loss orders or cause the execution of a Limit or Market order. Time Stops are flexible and very much strategy specific.

    ________________

    I will post a Learning Example showing how all these stops may be applied to a trade.

    Comments are welcome.

    Thanks,

    Chabah
     
    #30     Oct 4, 2006