PA is traded on NYMEX, not COMEX.
https://www.cmegroup.com/markets/products.html#exch=COMEX&sortDirection=desc&sortField=oi
Fully automated futures trading
- Thread starter globalarbtrader
- Start date
PA is traded on NYMEX, not COMEX.
https://www.cmegroup.com/markets/products.html#exch=COMEX&sortDirection=desc&sortField=oi
Oh, I see, so "COMEX listed metals (previously reflected as NYMEX) will be updated to exchange ‘COMEX’ " means that only metals which were ?incorrectly? labeled by IB as NYMEX will be changed to COMEX and those that are truly NYMEX like PA will remain NYMEX..
Because at the end the last email also said "NYMEX: All remaining "Metal" products" so I thought all NYMEX metals will be changed to COMEX..
Last edited:
Are you sure about "wave 4" being today? I had not received any notification in my email. And I see no such notification when I log in to the online account management.
See for yourself in TWS/IB app, on friday there was no COMEX/CBOT exchange names for any futures (except those mentioned in Wave 3).
I am surprised that IBKR has dropped the ball on this topic. I just restarted my trading software for the coming week, and indeed many instruments resulted in error messages. In my case I had to modify the exchange name for various grains, most (not all) metals, US bonds and the Dow micro (MYM) instruments.
I am surprised that they implemented these changes without informing their customers.
I see the same thing, and here's my stab at an explanation. Rob's forecast scalars are calculated across all instruments. But the average carry forecast can vary widely among different instruments because of the properties of the underlying commodities.
Just guessing, but I'd say live cows are probably about one of the most expensive commodities to take delivery of because of the "storage" costs associated with a live animal. This makes for a large (negative) carry. And apparently it's enough to just peg the carry forecast at the limit most of the time.
I'll go out on a limb, and say this is probably what you'd want, similar to how Rob used to have a rule for VIX that was just -10 all the time. It's only one forecast (or set of forecasts) out of many, with maybe a 20-25% overall weighting if you run similar forecast weights to Rob. Even if not ideal, it doesn't seem like it would do much harm.
In fact, this seems almost like a feature rather than a bug. Carry is determined to a large extent by the properties of the commodity, and in cases where the commodity has a large built-in carry, you'd want to bias forecasts in one direction. That is exactly what is happening here.
Last edited:
That makes sense,
although, I just checked Lean Hogs, which same as cows should be really messy to "store", probably even more
and carry on that instrument was hovering around zero\slightly positive over the last year.
Of course it depends on the particular price and carry contracts, and I can see that for LE carry is almost always negative month over month, when for hogs it becomes positive between Jul 2023 and Dec 2023. (not sure if these hogs are also alive and not frozen or something..).
The other cow contract GF feeder cattle displays the same carry pattern as LE - almost always -20, so that is consistent..
although, I just checked Lean Hogs, which same as cows should be really messy to "store", probably even more
and carry on that instrument was hovering around zero\slightly positive over the last year.Of course it depends on the particular price and carry contracts, and I can see that for LE carry is almost always negative month over month, when for hogs it becomes positive between Jul 2023 and Dec 2023. (not sure if these hogs are also alive and not frozen or something..).
The other cow contract GF feeder cattle displays the same carry pattern as LE - almost always -20, so that is consistent..
Last edited:
Couldn't have put it better myself.
If you think of this in terms of risk premia, basically when you trade carry you're earning a carry timing return (which you could strip out by demeaning the carry based on the historical mean of the signal), plus a systematic bias for a given asset class. The latter is probably 3/4 of the total SR of carry.
That makes sense,
although, I just checked Lean Hogs, which same as cows should be really messy to "store", probably even more
I'm no expert, but I think those poor little piggies are still alive (the rather macabre term on the CME website is 'ready for processing'...). But I guess pigs are smaller than cows, can be kept indoors, and have a lower food input per kilo of meat (which is one reason why beef is far worse for the environment than pork).
Rob
Thanks, but I bet you could have.
I just checked Lean Hogs, which same as cows should be really messy to "store", probably even more
A quick glance at the CME website shows the carry on cattle at around -10% YoY, whereas hogs are -1-2% (just eyeballing it), so there really is a big difference.
Another aspect to this is seasonal effects. That is probably why the carry on hogs changes direction occasionally (barring any unusual supply/demand imbalances).
Ideally, seasonal effects would be excluded from the carry forecast, because you would expect them to persist through time.
With carry, you would expect the futures price to converge toward the spot price (all else being equal).
With seasonal effects, you would expect the spot price to move toward the futures price, which isn't going to make you any money if you're holding the future.
I used to not trade carry on agricultural commodities for this reason. I brought it up to Rob in the past, and his conclusion was that it wasn't a big deal.
One way to exclude seasonal effects would be to calculate the carry forecast based on the 1-year away contract, rather than an adjacent contract.
I find myself curious about the asset allocation of the people on this thread. Is everyone well versed in "Smart Portfolios" and adhering to the recommended 25% maximum risk weighted allocation to managed futures? Or is there other strategies - like the outlook for stocks and bonds look awful, I'm going to take my chances on MF?