Hi Rob, just wanted to thank you for taking my question to TTU. I think you are right, I think I got a little obsessed with a few futures where this effect seemed more relevant. I don't have access to my system now but I will try to send some examples at some point. Thanks!
Another question on your excellent ST framework for rule development. I don't know if you have played around with non-linear/non-gaussian rules but this is what I am currently attempting. I am trying to backtest a couple of such rules but the issue I am facing is because of the non-normal behaviour they don't play nice with the usual forecast scaling (MAD or otherwise). I can still cap them but the mean abs is about 20-40 pct higher than the target resulting in an outsized influence relative to other rules 'on average'. The diversification benefits to portfolio in the backtest appear to be meaningful, but I wanted a second opinion on whether the inability to align the mean forecast with other rules make them undesirable or even dangerous inclusions in the portfolio.
So I finally tried backtesting of the short term mean reversion strategy (strategy 27) in my system using my own 1-minute bid-ask data I collected in the last ~4-6 years. But I used buffering of 4, 6 and 14 forecast-units (with the usual cap of 20, so only 1 step each way initially) to change positions (14 worked slightly better but not by much), i.e. I would only trade if the forecast changed by more than 14 units from the forecast that corresponds to the current position. And I used the same trading algo as for my trend system (no limit orders), but in backtests I simply assume I had to cross the spread on every trade. I tried 9 smallest instruments like US 2Y, MES MCG, etc. and I assumed I have full 300k capital for each of them, I also didn't reduce the base capital in drawdowns. After fixing many bugs (and I'm sure there's plenty more remaining), the results came out positive, i.e. it's not loosing money, but it's not making a lot either, the average single-instrument sharpe was 0.1 and for the overall 9-instrument-system 0.27. The average number of trades per instrument is 216, and the total number is 1945. Also the PnL curves look quite sporadic, I was expecting to see a classic negative skew pattern with small frequent gains and rare drops, and some instruments do show that, but not all. Also, there's no guarantee that my data is clean, the one protection I used (with the benefit of having historical daily prices) is to discard all bid-ask ticks which are outside of the Low-High of the current day (because my prod system does catch weird unrealistic prices during illiquid hours, and including them in such a backtest would inflate results). So I don't know, not a looser, but it requires capital, which would have to be subtracted from a system with the theoretical Sharpe of 1, so doesn't seem to make sense, unless I find some bugs in my backtest and it turns out to be better..
Relax, as long as: - they aren't a massive part of your portfolio OR - you have loads of them and they are relatively uncorrelated (central limit theoreom) ... it's fine Rob
Is this level of Sharpe given the number of trades statistically significant? I suspect it isn't. I've done similar tests (more trades but less data) and haven't found anything tradeable, not even as "informed trading with short-term signals for long-term investors."
Is it possible to test a simple filter using pysystemtrade: do trades only if they have the same sign as the ewmac (1,4) rule forecast?
It's done now. The additional 110 instruments (now totalling 215 liquid instruments) resulted in a sharpe increase of 0.15 in the backtest, which starts 1975. Not too bad but also not great. Interestingly, the trend-type strategies improved much more than the carry-type variants.
yeah, I agree, I wouldn't fully trust that it has a positive expectation based on only 4 years of data and 9 instruments (not even mentioning my potentially bad data quality), even among them there's no clear pattern and some of them are consistently loosing money. And with such a large threshold(step) the speed of the strategy is reduced substantially, so for a high confidence it needs more tata compared to something that trades every day..