horribillicus, you had changed your argument to initiate trades everyday to every Monday. you didn't specify holdtime, I was still assuming a one day hold time. fine, looks like it's a week hold time... blowout the first day should be changed to the first week. otherwise, this doesn't change anything. your example using the SPY's is clever and I would agree with you that the odds of it going to 0 in a week are near nil, but you never specified a market. What if you were short internet stocks in 97-98, etc or long some fraudulent penny stock? Since we have no knowledge of the market we're trading how can it be assumed that the risk of blowout does not exist and is not much greater than nill? Even if we did come up with some probability function using all the markets and stocks that have ever existed and figure out the % of which ones doubled or went to 0 at any given time and used that in the equation, the fact that yolanda's probability of experiencing a worse dd during that time does not change. her risk profile is greater, i don't know how else to express it. kevin, what are you not understanding? in this situation, with no bankroll or payout specified, we have to assume the avg std dev of returns is the avg risk/reward per trade. this simplifies the ror equation to basic coin flips in a row scenario. perhaps you need to be more specific. was it the long short exposure with an unspecified market that threw you off? read the above for more insight. man, you got it right, that's exaclty what the equation does.
I still have not seen these scenarios you speak of. Perhaps if you add enough qualifiers to your initial example you can find an extrmely specific scenario where it would be a bad idea. However, the resulting constrained scope would hardly allow it to backup your very broad statements above. Finally, I would agree with you on one point, "More is not always better". However, I would like to add a qualifier of my own: just 99.999% of the time it is..
ill add my two cents. I trade 8 different strategies. the best strategy has a profit to drawdown ratio of about 13 to one. when i add all my systems together, total drawdown does not increase but profit does and my ratio of profit to drawdown increases to like 46 to 1. correlation is an afterthought. i will add a system to my portfoliio if it increases the total profit to drawdown ratio. thats MY bottom line. profit to drawdown ratio.
totally agree. no matter what return/risk ratio take, that is what the whole multiple system thing is about. this is pro thinking. no offense intended but everything else is in my eyes amateurish or wishful thinking (MY system will never die, this kind of hope ...).
"I trade 8 different strategies. " Are they all trading the same instrument? "the best strategy has a profit to drawdown ratio of about 13 to one." Are your profit and drawdown expressed in absolute dollar terms? If so, I've found that this ratio is very much dependent on the length of time considered. What time frame did you use? "correlation is an afterthought. " Well, not really. The reason you're getting such high profit to drawdown ratio is that the strategies are not highly correlated. Ideally you want negatively correlated strategies. "thats MY bottom line. profit to drawdown ratio." Same here. After all we want the most bang for buck with minimal pain. Richard
practically negative correlated strategies are the exemption. corrs between 0.1 and 0.3 are quite enough. in addition i would recommend a concept i call stressCorrelation which is asymetric between two time series and just looks at stress events (=down days for example) an the relation to the other time series. the idea is that you do not care about positive corr when things go north in both systems.
There's a pretty easy way to create very negatively correlated strategies: by design, ensure that whenever strategy A is Long, strategy B can be Short or Out, but it cannot possibly be Long. Ensure that A and B usually have opposite positions, and never have the same position. The most straightforward embodiment of this idea is: create a filter which indicates whether the very long term trend is up or down. MACD(300,50,3) is one example of such a filter. Strategy A is only allowed to accept new entry signals in the direction of the filter (when filter says "down", strategy A can enter Short but it cannot enter Long). Strategy B is only allowed to accept new entry signals in the opposite direction of the filter (when filter says "down", Strategy B can enter Long but it cannot enter Short). Voila, strategies A and B cannot both enter long (or both enter short) at the same time. Thus their correlation coefficient is somewhere between -1 and 0. In practice I routinely see correlation coefficients between -0.4 and -0.7. In practice it is desirable for strategy B, the one that trades opposite to the direction of the very long term trend, to be a relatively short term, "opportunistic" strategy: observe a setup, get in, make some money, and quickly get back out. Exploit short term contra moves while swimming upstream against the very long term trend. Strategy A is much more forgiving, since it trades in the direction of the very long term trend. You have a lot more choices here. You can add extra logic if you like, to deepen the negative correlation some more. You can make the entry conditions for Strategy A yet more restrictive, from Code: If (A has an enter long signal) AND (the filter says up) THEN Buy to Code: If (A has an enter long signal) AND (the filter says up) AND (strategy B is not long) THEN Buy Naturally you'd add a similar third condition to the entry logic of strategy B. This modification handles situations where the filter switches states before a trade exits. Finally, you can incorporate knowledge of B's position and intentions, into A's exit logic, and vice versa. A simplistic form of this "manufactured negative correlation" approach is used every day in the managed money arena: let A be a "long only" strategy, and let B be a "short only" strategy. Presto, negative correlation. There are hedge fund managers who trade only from the short side; asset allocaters blend them into a mix with other long-bias managers, and reap the benefits of deep negative correlation.
i admit i never thought of that and intentionally develop a negative correlated system. maybe i should. your post was quite inspiring in this respect. my problem is that we manage several million and this requires some market diversification as well, so it is not so easily done like you said, but nevertheless, worth some thoughts: thnx!
I do it on a portfolio of more than 50 commodity markets. Admittedly some of them are a bit thin (volume of front month less than 10,000 contracts per day) so it becomes slightly worrisome if you're trading more than 250-500 contracts per ticket. If you risk 1% of equity per trade, and your stop is such that 1R = $500, then a $12.5 million account would be trading 250 contracts per ticket, which is not unthinkable. In my case, I risk considerably less than 1% of equity per trade, and I use pretty wide stops so my 1R value is on the average a lot more than $500. Therefore I don't reach 250 contracts per ticket until well beyond $50 million.