If you're trading forwards or futures, be sure to include the costs of continuously rolling over your positions: you pay the bid-ask spread AND the commissions, on each and every roll. I'll mention the fourth-best and the third-best ways to size positions in this context, and leave it to your own ingenuity and experimentation to find the best and second best methods: 4. Trust that your broker and the exchange have set the margin requirement to a reasonable value. Then positionsize = K * YourAccountValue / MarginRequirement . You choose a value for K that suits your personal risk tolerance profile. 3. Trade one share / contract / parcel, for every D dollars in your account. Positionsize = (1/D) * YourAccountValue. You choose a value for D that suits your personal risk tolerance profile.
Hi achilles, Your contribution is the only thing worthwhile in this whole junk thread. As jackbird said "fuzzy maths" only useful for losers and their math quack sycophants.. Don't forget that dadioos like van Tharp only cater to that kind of greedy all-believing public. Of course if you insist, MM makes sense if you understand this as your own global technique of "how to speculate with your money". No little formulas apply in any general way to this endeavor. Win or loss depends on the judicious manipulation of quite subtile insights. Mathematics can have its place in this but almost nothing can be naively borrowed from probability theory alledgedly applied in a haphazard manner by doctrinaire know-nothing losers.
nononsense, this is the experience of one of the Kelly criterion users ... Quote from Ed Thorp How does the Kelly-optimal approach do in practice in the securities markets? In a little-Known paper (Thorp,1971) I discussed the use of the Kelly criterion for portfolio management. Page 220 mentions that 'On November 3, 1969, a private institutional investor decided to ... use the Kelly criterion to allocate its assets.' This was actually a private limited partnership, specializing in convertible hedging, which I managed. A notable competitor at the time (see Institutional Investor, 1998) was future Nobel prize winner Harry Markowitz. After 20 months, our record as cited was a gain of 39.9% versus a gain for the Dow Jones Industrial Average of +4.2%. Markowitz dropped out after a couple of years, but we liked our results and persisted. What would the future bring? It is now May, 1998, twenty eight and a half years since the investment program began. The partnership and its continuations have compounded at approximately 20% annually with a standard deviation of about 6% and approximately zero correlation with the market ('market neutral'). Ten thousand dollars would, tax exempt, now be worth 18 million dollars. To help persuade you that this may not be luck, I estimate that during this period I have made about 80 billion worth of perchases and sales ('action', in casino language) for my investors. This breaks down into something like one and a quarter million individual 'bets' averaging about 65,000 $ each, with on average hundreds of 'possitions' in place at any one time. Over all, it would seem to be a moderately 'long run' with a high probability that the excess performance is more than chance.
Just 2 cents: When people can freely and arbitrarily choose a fraction (say 1/2, 1/3, 1/4, etc.) of an optimised whole Kelly value in order to derive a supposingly suitable position size to fit an individual's subjective purposes/ needs, mathematically and logically people would Not need Kelly anymore. However, I believe Kelly would be still good and useful for some MM vendors to promote and market their trading services, just like many TA vendors do.
If you know and trust enough your system, you should risk the Kelly fraction and rip the maximum return. If you can't stomach a large drawdown, you have the tool to determine the risk that will keep it under your acceptable level, with the probability you want. If you don't know or don't trust your system, it means that you don't trade, you just gamble. Arbitrarily risking 1%, 2%, 6% of your account might look like a wise choice, but actually you don't know if you leave money on the table, or if your risk is above the top (you'd get the same return with a smaller risk, on the other side of the Kelly curve). Money management means position sizing, and it is independent of exits and entries. Ignoring it means you'll not get as much as you could from your trading system, and could even ruin a profitable system.
Looks like you after starting this thread have now trusted your knowledge of MM very much - perhaps good enough to provide MM consulting services. Good luck again to your journey.
No need for consulting services. It’s all there by Ed Thorp himself. http://money-management.martinsewell.com/Thor97.htm All you need is a few days to adjust this framework to your particular trading systems portfolio. It will help you to 1. Determine what are the best systems to trade 2. What is the appropriate capital allocation between them and what is the expected time to reach your goal 3. What must be your exposure for a preferred expected maximum drawdown A relative good math background is required. Good luck
You're right: from the time I started this thread I reviewed several MM papers, played and charted formulae, and now I have a fresh and clearer picture of how to size my positions. This thread benefited me, and hopefully a few others. Yet I'm not an expert in this field, and anyway I'm not interested in consulting. Recently I also started reviewing my stop-loss and exit strategies.
What is the rationale for using Kelly as I thought it was for Bernoulli distributions only, not the typical distributions you get in trading? DS