Max Sharpe ratio https://github.com/robcarver17/pysystemtrade/blob/master/syscore/optimisation.py Line 525 GAT
Hi GAT - I made a first, rough attempt at implementing the EWMAC strategy described in your book on ES-minis. I had a quick question which is really just a sanity check to make sure I've understood the moving parts... Using different variants of MAs, I only start to get positive performance using the relatively longer pairs, e.g. 32 / 128 days. Is this consistent with your results for this contract? I'm reluctant to play around with too many pairs as I don't want to introduce overfitting (too late!) but I'd also like to make sure this is due to spx being too noisy for short MAs rather than an error on my part (also quite possible). Thanks.
It's probably not an error on your part. Fast trend following doesn't seem to work so well on equity indices (though it's flat, rather than money losing) at least after the mid 1990's. However would this justify not using them? Well 'doesn't work so well' isn't a statistically significant difference. In case you're interested here are my own weights. Only the fast lookback is shown. momentum4 0.4 momentum8 0.32 momentum16 0.08 momentum32 0.06 momentum64 0.14 Splitting your portfolio across each of the three slowest speeds wouldn't be a stupid thing to do, but I wouldn't be more extreme than that. Rob
Rob could you please let me know the current forecast you have for carry on euro stoxx? I'm trying to understand why you have such a short position when I see a moderately strong carry forecast. I suspect something may be wrong in my implementation. Thanks!
Another milestone yesterday, new HWM set at +100.7% at lunchtime. Over 6% up on the day. In other words if I'd maintained the same risk capital throughout I would have doubled my money (I've actually made about 132% on my current capital at risk and 176% on the original 300K) GAT
Congrats to those results. Question about "risk": I can estimate individual position risk using the instruments volatility as a proxy: risk = position_value * stddev(percent_instrument_return) But how do you estimate the total account risk? Just adding up all individual position risk values gives a too high number, essentially a worst case risk. Is the sum of all position risk values divided by the Instrument Diversification Multiplier a good estimate? Would using the account volatility under-estimate the risk?