Determining Scalability: Seeking Guidance

Discussion in 'Risk Management' started by btowntrader54, Jul 24, 2008.

  1. Hi All,

    I would appreciate some advice/guidance. I've developed a trading indicator for use in intraday positions. There's a maximum of 1 buy and 1 sell per day per equity.

    All fall under these criteria.

    Min Mkt Cap: $1b
    30d avg vol: 300k
    Price > $15

    Any advice on how to determine the maximum size I should be taking per trade based on the minimum crits?

  2. Tums


    first learn to trade, then learn to scale.

    p.s. this simple answer has more meaning than it appears.
  3. In terms of constructive advice that doesn't qualify.

    Let me try this. How does one determine scalability? Based on the information above shouldn't I be able to figure out the ideal maximum share count based on those minimums (the level where market impact will be minimal)?
  4. nassau


    It will first be determined on your capital and how many trades you are looking at doing daily as perservation of capital should be your no. one rule
    I would then look at beta as the volitility will also be an influence in size and the average behaviour ie does a few shares gap or cause a retracement
    the sentiment and direction ie in a channel, high, low
    short interest

    and a clear exit and risk management

  5. Nassau,

    I appreciate the response.
  6. MGJ


    Load up some 1 minute bars (which might be the appropriate granularity, but use your own judgment based on the system's characteristics). For each bar, plot a dot on a scatterplot. Let X = (volume of this bar) and let Y = (price change, from last bar to this bar). Use your own judgment to decide whether price change means max((Hprev - Lthis), (Hthis - Lprev)) or abs(((Hprev - Lprev)/2) - ((Hthis - Lthis)/2)), or something else.

    Examine the data to see whether it supports the hypothesis "large volume causes big price changes". If so, assume that "big price change" caused by "large volume" is slippage on your entry and on your exit. How much slippage can your system tolerate and still spray a monstrous fire hose of profits into your account? Wherever your "Ouch" point is for slippage, that's how much "big price change" you can tolerate. Find the corresponding volume (on the scatterplot) and that's your maximum trade size, the upper limit of your scaling exercise.
  7. You're an animal MG, thanks
  8. for longer term trading (not intraday)


    Position Size in Dollars=

    (1-3% of your capital)*(C / 3*atr(10))

    you can use a longer timeframe for volatility, or use StDev instead of ATR.

    You just add a factor below to the system to limit the size in stocks or futures that aren't very liquid.

    You can limit your size to say 10% of the daily volume.

    Position Size in Dollars = Min((C*Volume)*.1, PositionSize)

    This will allow you to avoid trading too big in a small cap or small futures contract.
  9. If I recall correctly the rule of thumb is not to use more than 0.1% daily stock volume for your position.
  10. You can use whatever number you want.
    #10     Jul 24, 2008