Search results

1. isn't it only for stocks? anyway, I never bothered with that (maybe unwisely?) 2. I think pysystemtrade's roll calendar is a good start, but then in my system for "hard" and "soft" roll periods I refined overlapping periods with maximum volumes for previous and next contracts for each...

I use forecast grouping from the Rob's last book, 60\40 divergent\convergent, then mostly equal weights within them, although I think I gave skew a little less weight than carry..

IB has data quality issues (-1, 0, incorrect prices, it's been mentioned here before), not too often, and I think mostly temporary, i.e. IB will eventually correct a wrong price in the EOD history.. Actually I often see lots of price warnings from my system on Monday after holidays\long...

I think with handcrafting the top level weight allocation shouldn't really be 1\n, exactly because some groups have too few instruments in them (or we should try to come up with some even higher-level groups e.g. 'financials' for all stocks, bonds and volatility).. And on the other hand, the...

I used to run full DO on every tick on ~100 instrument portfolio, but my server was choking on that and I didn't feel like upgrading it for really no reason (running it that frequently isn't adding any value), so now I'm running it every 10 seconds (implemented in C#), the load on all 6 CPU...

Our tracking errors could be different because mine is "daily" and maybe yours is "annual", I feed daily standard deviations into the calculations, if I convert it to annual (0.359*16=5.7%) it becomes much closer to yours.. (also, compared to my regular target daily risk, which is 4687.5$, my...

hmm, interesting.. I reran my backtest, there was a small error, but the result didn't change much, still the number of trades and margin\notional usage gets ~35% higher, when starting from the rounded weights... You're not seining an increase in the number of trades\capital usage? I run this...

Btw, how do you calculate tracking error in percentage? I normally calculate tracking error in currency, e.g. an actual value could be 1,077$, how would you convert it to % ? I see 2 options: - divide 1077$ by my system's base capital (e.g. 300,000$) - or divide 1077$ by the total notional cost...

Also, it seems that my margin usage and net exposure went up somewhat (I'll double check it just in case..): original DO with starting zero weights (margin usage then notional exposure): New DO where with starting rounded ideal weights:

I've just ran a quick backtest where DO starts with the ideal rounded positions instead of zeros, the PnL and Sharpe stayed roughly the same, but the number of trades jumped from 6558 to 8631, so by 30%.. The overall costs also jumped, but apparently a slightly higher return compensated for it...

I'm also trading this one, but never had an actual position in it, probably that's why IB didn't send me this email, thanks for sharing!

So what logic is currently included into the automatic rule-weight calculation, apart from dropping expensive rules? In the Strategy report "Forecast weights" I can see that e.g. SP500_micro (a cheap to trade instrument) has breakout20=1.2 but breakout160=NaN, so a more expensive variation is...

Here it is, some weights are 0, or set to 'don't trade' for various reasons, e.g. repeating instruments with different multipliers.. There's probably more liquid instruments now, I need to re-check and add them, e.g. the "new" lumber, but the process is laborious and I just can't bring myself to...

The last couple of months were good for me too, not at HWM yet, but if things go the same way for a couple more months I might be..

Intuitively, though, it still seems strange that we're considering trading costs if we're not actually doing any trading with those very expensive instruments, and we can't even use them for forecasting.. I.e. why the expensiveness of an instrument also makes it bad at making forecasts.. But it...

The way I remember this post's conclusion is that basically throwing all instruments with all the rules into the DO and letting it automatically select\penalize expensive forecasts\instruments isn't optimal. But I didn't quite get the nuance about what should happen with the instruments which...

I've been thinking, so when some rules are too expensive to trade for an instrument, we assign 0 weight to those rules for that instrument, and if all rules are too expensive, we remove the whole instrument., But with DO, when an instrument is too expensive, we still keep it, just set maximum...

https://www.cmegroup.com/markets/interest-rates/us-treasury/3-year-us-treasury-note.calendar.html I'm not trading this one, but on CME it says FIRST NOTICE\FIRST POSITION\FIRST HOLDING for March (last trade 28 MAR) is around Feb 26, so I'd try to be out of that contract probably by Feb 20-24...

I finally tried the 70\30 volatility estimation in my system (only took me a couple years :) ) These are 'Australian Dollar Futures' and 'US 10y', orange is the new way of estimating with 30% of average vol over 10y +70% of the recent vol, the blue line is 100% of the recent vol as before. So...

yeah, that's probably the best, but it will require more changes in my system, so at first I'll just try it with my current execution algo, but the signal will be generated not using the last\mid price but the bid or ask price (whichever is 'worse').. (my system is event-based so I'm processing...