I find it puzzling that breakout didn't do it for you when looking at this graph. I wonder what can explain the discrepancy?
I think this chart could be easily misinterpreted. The performance is very strong for the first five years. Rebase the graph at 1990, what would you see? Secondly, these lines don't look they've had the volatility normalised, so I'm not sure it makes sense to compare them on a chart like this. (@globalarbtrader care to help out here?) And lastly, with respect to the shape of the breakout curve- the curve with the highest sharpe ratio will be an almost flat slope up and to the right (the strong performance at the beginning might lower the sharpe). In any case, perhaps I missed something, so I will go back and test again
Hmmm, maybe I'm thinking of this incorrectly. If a backtested sharpe of .92 is not statistically significantly different from .86, then I would think I should add the rule. My logic is: since I can be fairly certain that it will add a bit of diversification to my trading rules, and it doesn't substantially impact my backtested sharpe, it should provide a small bump to my actual sharpe. Does this make sense?
Eurodollar about the same as you. v2tx 3 contracts out, too volatile at the front end for me. Are you still running the crypto market maker? What are your thoughts for next research? I'm evaluating if there are any options strategies I could run using daily data, or rather if there are any strategies which are suitable for budget hardware and a part time portfolio manager.
It's stuff that takes advantage of the market micro-structure in one or another form. Frequently, you'd find stuff that would not be profitable on it's own (e.g. round trip commissions would make it not interesting to players that have better tools) but since you have a longer-term strategy, you can use it as an overlay.
LOL, it's hilarious that I was just answering a question in the options section that was more or less dealing with "verification periods" for S/R. Here is a snippet of code that does the opposite - gives you a table of how much time you need to prove to yourself that a strategy's SR is stable: Code: ntp = lambda sr, ci: norm.ppf(ci)*norm.ppf(ci)*(1+sr*sr)/(sr*sr) SR = np.arange(0.5,5.0, 0.5) CI = np.array([0.9,0.95,0.98]) print pd.DataFrame([[int(round(ntp(x/16, c)/21)) for c in CI] for x in SR], index=SR, columns=CI*100) PS. this one assumes annualized S/R based on daily P&Ls
As a person that trades systematic volatility strategies for a living, I'd say that (a) pure vol strategies are very capital intensive, even with PM or SPAN (b) options are a pain unless you can execute by hand at least some time (c) if you have good directional ideas, it's better to trade them in delta-1 (d) there are interesting things to do in vol if you know where to look
I think once you approach a Sharpe of ~1, there isn't much you can do to really improve things on a daily strategy (correct me if I'm wrong!), even if you wrote a whole extra strategy and diversified into it (e.g. AHL Dimension). For the most part, I'm just leaving it alone. At some point I'd like to write rules specifically for energy trading just for fun. I'm writing market making software for cryptocurrencies, targeting a Sharpe of 4. I'm trying to start a company to do this.
You have the backtest for a Sharpe 4 strategy? It is possible to achieve higher than Sharpe 2 on daily or even weekly strategies if you work with equities (long short stat arb). The universe is much larger so this much more diversification than just a few futures
Equities don't have much volatility nowadays. Also borrowing costs is a killer. There are other hidden costs like taxes on dividends received but nothing on taxes paid. Are you still able to achieve sharpe 2 these couple of years?