[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: US T-Bond, 30 year.



PureBytes Links

Trading Reference Links

Hello Adam,
It would be a simple thing to exit all contracts at the end of 99 and
buy the same back in 2000, that would avoid giving you a false profit,
the little higher cost in commissions shouldn't upset a good system.

best regards
foolsgold

AH> In 2000 the US 30 year T-Bond future changed its spec and the data I 
AH> have from CRB Trader was adjusted to compensate.

AH> CRB Trader say basically the spec was changed from 8% to 6% for the 
AH> notional underlying bond.

AH> Unfortunately the pre-2000 data is not back-adjusted, it's all just 
AH> nearby-futures with no adjustment at roll-over, all in one file.

AH> What CRB Trader did with their historical data was to lump all the 
AH> pre-2000 data into the one file and call it contract 1999Z. This data 
AH> came direct from CBOT and CRB Trader didn't change it, apparently.

AH> I have the old original pre-2000 contracts still, and am compiling a 
AH> back-adjusted continuous contract.

AH> However when comparing mine against the CBOT data, I immediately see 
AH> price differences, sometimes between days, sometimes relatively between 
AH> one day's OHLC points and I'm worried about back-testing against it.

AH> My instinct is not to worry about it. Surely the back-adjustment should 
AH> iron out price differences due to the spec change, even if in some of my 
AH> tests, I have a trade appear which spans that 1999 / 2000 roll-over?

AH> Does anyone disagree?



AH> Adam