Fully automated futures trading

Discussion in 'Journals' started by globalarbtrader, Feb 11, 2015.

  1. Quick question regarding minimum capital calculations for the starter system in Leveraged Trading please -

    The minimum units I can buy on CMC (CFD per point) is 0.02, this currently equates to an amount of £256 (see pic below) - So is this the minimum capital required?

    I only ask because when I do the formula for it I’m getting a different amount -

    Minimum capital = (minimum exposure x inst risk) / Target risk

    So for the Nasdaq (Dec 2023) Forward on CMC I’d first find my minimum exposure :

    Minimum exposure = 0.02 x $15,873 / 1 = $317.46

    Then :

    Minimum capital = ( $317.46 x 19.12%) / 12% = $506 / £408

    Many thanks

    [​IMG]
     
    #3851     Nov 14, 2023
  2. No that's the minimum exposure which you have used in your formula already. Why would CMC know about the concept of minimum capital and even if it did, how could they know what your risk target is???

    Rob
     
    #3852     Nov 14, 2023
  3. Oh damn it yes! sorry i see that now! Ugh sorry about that. Long day, head has not left the book!
     
    #3853     Nov 14, 2023
  4. Hello Rob,

    I hope you're doing well! This thread is gold. :fistbump: I'm currently diving into Strategy 26, but I've hit a bit of a snag. Could you lend a hand in straightening things out for me?

    When it comes to getting risk-adjusted forecasts from Strategy 26, I'm a bit unsure about the σp,t calculation.
    Should I be using the formula from Strategy 1 which appears after the 'why not Draw down' column, particularly the standard deviation part? Or is it better to go with the 'All other directional strategies position sizing' from Appendix B?
     
    #3854     Nov 15, 2023
  5. Neither

    It's a daily standard deviation of price returns

    So either use price differences to estimate directly or multiply the standard percentage vol estimate used from Strategy 3 onwards by the current price

    This is literally what it says in the book I am not telling you anything new here....
    Rob
     
    #3855     Nov 16, 2023
    No.Emotions.Trader likes this.
  6. AlexCh

    AlexCh

    Hi, Rob!

    As you blogged repeatedly in your April annual reports, you have a long only portfolio of stocks and ETFs in addition to systematic futures trading.

    What do you think about hedging this portfolio by short selling of index futures on some trend signal? As I know you do not hedge.

    I understand that there are index contracts in your futures system, but they are not related to the size or volatility of a long portfolio.
     
    #3856     Nov 16, 2023
  7. :banghead: ah crap I wrote the wrong symbol and totally out of zone.
    I somehow thought words Part One from p.423 as Strategy One....:banghead::banghead::banghead:
    I was wondering myself, "why the heck Rob is talking Stdv from strategy one in strategy 26" and now all things are cleared up.

    I guess 'use price differences to estimate directly' is Risk adjusted forecasts, daily price standard deviation from p.604 and 'multiply the standard percentage vol estimate used from Strategy 3' is σblend,t from p.80. Right? :sneaky:

    I hope I'm on the right track:strong:

    Thanks Rob!
     
    #3857     Nov 16, 2023
  8. That would increase the risk allocation of my portfolio to momentum....

    Rob
     
    #3858     Nov 16, 2023
  9. Kernfusion

    Kernfusion

    I was thinking.. DO takes costs into accountant therefore will probably never actually trade expensive contracts, even if they have great trends, i.e. it will always try to substitute them with something cheaper, even if the forecasts of those other things aren't as great.
    Is that a good thing?

    E.g. I was looking at Orange Juice contract in my system, it had a great upward trend and the overall combined signal was +20 for the better part of the year, yet the system never took a position in it.. It's current median bid-ask spread is 90USD which is quite high (although in my outdated settings it's 22USD, but it's still higher than 50% of all instruments).

    There could've been other things that prevented the system form taking positions in Orange Juice of course, e.g. low instrument weight, correlations and the forecast levels of other instruments.. But maybe something else, e.g. my ShadowCost parameter is too high.. (which I don't think it is, it's just ~2.5 at he the moment)..

    I also checked across all instruments, and it is indeed the case that the first half of my instruments, sorted by the median bid-ask spread (high-to-low) have zero or only a few trades in all the life-trading history (which is only several years), with the exception of VIX, Swedish kroan, US20 and Wheat..

    Maybe all of this is fine and what we actually want the DO to do, it just seems that some well-trending but expensive instruments are being ignored..
     
    #3859     Nov 17, 2023
    newbunch likes this.
  10. Would it then make sense to see if you can make your order execution better? Meaning that you find a fill in between the bid and ask price?
     
    #3860     Nov 17, 2023