Holding a rates future is not the same as betting on the underlying spot rate. You earn the futures roll.
It's the same reason why it could still be profitable to short vix futures when vix drops under 10 and could only go up from there.
Fully automated futures trading
- Thread starter globalarbtrader
- Start date
Holding a rates future is not the same as betting on the underlying spot rate. You earn the futures roll.
It's the same reason why it could still be profitable to short vix futures when vix drops under 10 and could only go up from there.
It sounds to me that you are having a bout of "fear of high altitude". With smaller contracts may the position size increase to levels you have not seen previously and which may cause concern. However, after having reached these levels of position size several times you get used to it.
On the other hand: having a large position size gives the system the opportunity to track the forecast (and volatility) more precisely. Resulting in less "quantization noise", which is of benefit.
I guess you're right..
Also, if I look further back, this volatility isn't unusual, it was almost as low just 2 years ago:
Futures price is not the same as spot, but it's still related right?, so in order for it to go UP (and my long position make money), the 2y US interest rate (or the future expectation of it) must go DOWN, and I just don't see how it can happen, there's almost no room left for it (apart from negative rates)..
Roll returns is carry signal, and it's currently almost zero in my system, slightly negative. Just checked, my price contract is '2021-03' and carry is '2021-06' the prices of these 2 contracts at this moment are exactly the same 110'15'2, so no return to be earned there:
Although, I'm not measuring carry from just one single point, it goes back and might even involve prices from an older pair, and then I smooth it and I have 3 permutations with different look-backs, yada yada... but even so, the current combined carry signal on this instrument is slightly negative, which I guess means that the system isn't expecting to earn the return from carry but from trend signals, mostly these with long look-backs, which still remember the rise in price several moths ago..
I mean it's not something particularly horrible (getting bored, I guess, and want to "improve" something
), and I'll probably let it be as indeed these contracts are small, but still, from the logical perspective not sure if it makes sense to keep holding them right now..
Also, if I look further back, this volatility isn't unusual, it was almost as low just 2 years ago:
Futures price is not the same as spot, but it's still related right?, so in order for it to go UP (and my long position make money), the 2y US interest rate (or the future expectation of it) must go DOWN, and I just don't see how it can happen, there's almost no room left for it (apart from negative rates)..
Roll returns is carry signal, and it's currently almost zero in my system, slightly negative. Just checked, my price contract is '2021-03' and carry is '2021-06' the prices of these 2 contracts at this moment are exactly the same 110'15'2, so no return to be earned there:
Although, I'm not measuring carry from just one single point, it goes back and might even involve prices from an older pair, and then I smooth it and I have 3 permutations with different look-backs, yada yada... but even so, the current combined carry signal on this instrument is slightly negative, which I guess means that the system isn't expecting to earn the return from carry but from trend signals, mostly these with long look-backs, which still remember the rise in price several moths ago..
I mean it's not something particularly horrible (getting bored, I guess, and want to "improve" something
), and I'll probably let it be as indeed these contracts are small, but still, from the logical perspective not sure if it makes sense to keep holding them right now..
Last edited:
I guess you're right..
Also, if I look further back, this volatility isn't unusual, it was almost as low just 2 years ago:
View attachment 247750
Futures price is not the same as spot, but it's still related right?, so in order for it to go UP (and my long position make money), the 2y US interest rate (or the future expectation of it) must go DOWN, and I just don't see how it can happen, there's almost no room left for it (apart from negative rates)..
Roll returns is carry signal, and it's currently almost zero in my system, slightly negative. Just checked, my price contract is '2021-03' and carry is '2021-06' the prices of these 2 contracts at this moment are exactly the same 110'15'2, so no return to be earned there:
View attachment 247751
Although, I'm not measuring carry from just one single point, it goes back and might even involve prices from an older pair, and then I smooth it and I have 3 permutations with different look-backs, yada yada... but even so, the current combined carry signal on this instrument is slightly negative, which I guess means that the system isn't expecting to earn the return from carry but from trend signals, mostly these with long look-backs, which still remember the rise in price several moths ago..
I mean it's not something particularly horrible (getting bored, I guess, and want to "improve" something
This is a debate I've had long and hard, with other people and myself. And I had this debate.... in 2011. When Swiss LIBOR contracts started to price negative. I spent ages building a system that degeared the position as rates went to zero (which in the code went by the variable name 'the singularity'). I hope they got rid of that system after 2013. It would have closed many profitable European bond positions over the last few years. It goes without saying I don't have such a system in place right now.
Yes, you are betting on negative interest rates. But your system doesn't know that, or need to know that. We know that negative interest rates aren't impossible (we didn't, in my defense, know this in 2011). I mean, negative oil futures prices aren't impossible it turns out. It's always dangerous to assume anything.
Low vol is another story. I stopped trading SHATZ for that exact reason. But as you say, the vol has been lower before, so you can't use that excuse here.
It might be worth putting in a hard per instrument leverage limit, or a 'black swan' limit that limits position size to something that would prevent you from losing x% if the market fell by the largest one day move ever. Something quantitative that applies to all instruments.... not just a hack because you don't like the position!
GAT
Yeah, that makes sense, just wanted to make sure I understand what's happening..
Btw, negative prices would really be a problem for me, as this is one of the few real-time price checks I have - the system will ignore RT prices < 0., and I also recently added a filter in the EOD price collection routine to ignore zero prices (because IB keeps sending me a clearly wrong closing price for CAC40 exp:2020-09-25 newP.TS:2020-09-21 12:00:00 AM, which was throwing off my forecasts).. I guess I'll allow negative prices (but will still ignore zero ones because an absolute 0.0 price would still be rare in real life and should be ok to ignore..). Doing it with a standard deviation range is probably a better way though..
Also, my PROD system just turned slightly positive for the year, looks like ZS, ZC and AUD made some money, so things are good (at least at the moment
)Hey Team,
It's been a while since I logged into this forum. Great to see KernFusion and HobbyTrader still here, as well as (and of course) GAT.
I had a great year at +37% with my own systematic strategy, where I've taken bits of GATS' amazing framework and added a lot of my own adjustments. Needless to say, it was a gut-wrenching year and even though I made good returns, it was exhausting to watch...
Setting the slate clean now for 2021 - hat's off to you all and good luck trading in 2021.
It's been a while since I logged into this forum. Great to see KernFusion and HobbyTrader still here, as well as (and of course) GAT.
I had a great year at +37% with my own systematic strategy, where I've taken bits of GATS' amazing framework and added a lot of my own adjustments. Needless to say, it was a gut-wrenching year and even though I made good returns, it was exhausting to watch...
Setting the slate clean now for 2021 - hat's off to you all and good luck trading in 2021.
That's a great performance. I'm guessing you have quite a high vol target, if it was 'gut wrenching'? The final performance for me was +12.5% for the calendar year. I made about 20% in my core futures trading, and lost about 8% in my ETF plus short stock hedge.
For the financial year so far, I'm basically flat.
My new trading PC is now happily whirring away behind me. It's so... fast. The daily backtest is down from 33 minutes to 3 minutes! Of course I could learn to write better code, but it's a hell of a lot easier just to buy a shiny new box...
GAT
Most of my net worth is tied up in a 10% vol risk-parity multi-asset class based strategy with couple of other strategies in volatility & options based trading. No futures based portfolio yet. Up 12% for the year in my home currency (SGD), max draw-down of 10% . Lost 3% to currency effect (Five figure sum which is painful to me as I'm on a sabbatical to expedite my CS masters in Georgia tech and dependent on my portfolio for cashflow haha. Brings me to the point, does anyone hedge their portfolio to their home currency here?).
Last edited:
Hey all, here are my stats for the year (actually, starting January 27th, since that was my first trade for the system).
It was not looking good in October, I'll tell you that. That was just over 30% drawdown. Risk target is 24%.
I wrote before about the spike in the beginning of the year, but to summarize, it was a combination of luck (biggest part), very few positions on (I started trading with 3 markets and am now at 20), and 2 bugs, one in carry calculation and one in risk management. But mostly luck.
If I was running the system and markets I'm running now for the whole year, it looks a lot less wild and less impressive.
Overall, I'm pretty happy since this year was mostly about learning for me - when I started this in January, I remember thinking I'd be happy to just cover the transaction costs and break even.
Thanks again to everybody here for all the help, especially Rob whom I still owe some pints, but that will have to wait a bit more.
It was not looking good in October, I'll tell you that. That was just over 30% drawdown. Risk target is 24%.
I wrote before about the spike in the beginning of the year, but to summarize, it was a combination of luck (biggest part), very few positions on (I started trading with 3 markets and am now at 20), and 2 bugs, one in carry calculation and one in risk management. But mostly luck.
If I was running the system and markets I'm running now for the whole year, it looks a lot less wild and less impressive.
Overall, I'm pretty happy since this year was mostly about learning for me - when I started this in January, I remember thinking I'd be happy to just cover the transaction costs and break even.
Thanks again to everybody here for all the help, especially Rob whom I still owe some pints, but that will have to wait a bit more.
Hey all, here are my stats for the year (actually, starting January 27th, since that was my first trade for the system).
View attachment 247956
It was not looking good in October, I'll tell you that. That was just over 30% drawdown. Risk target is 24%.
I wrote before about the spike in the beginning of the year, but to summarize, it was a combination of luck (biggest part), very few positions on (I started trading with 3 markets and am now at 20), and 2 bugs, one in carry calculation and one in risk management. But mostly luck.
If I was running the system and markets I'm running now for the whole year, it looks a lot less wild and less impressive.
View attachment 247958
Overall, I'm pretty happy since this year was mostly about learning for me - when I started this in January, I remember thinking I'd be happy to just cover the transaction costs and break even.
Thanks again to everybody here for all the help, especially Rob whom I still owe some pints, but that will have to wait a bit more.
Btw, @wopr which contracts did you decide to add? I'm also going to add more capital\products in a month or so and thinking what to pick.
currently I have these:
GE, MXP, ZC, V2TX, ZT, ZS, 3KTB, OAT, GBM, NG, ZW, LE, AUD
Want to add a long-term bond, probably BTP (which I badly missed this year
), a precious and a base metal, probably platinum and copper, maybe try adding micro Nasdaq(MNQ) or ESTX50, and maybe a currency e.g. GBP, also maybe crude.. Everything else is either too big or too correlated.. Was also thinking about ice-futures, like sugar\coffee\cotton and contracts from other exchanges like DJ600, EU600, XINA50, NIFTY, SGXNK - but these need more work for collecting historical data and implementing market orders as there's no way I'm paying 200$\month for real-time data from ICE