Risk/Position adjustment if the future's underlying changes?

Discussions about Money Management and Risk Control.
Post Reply
ADMP
Roundtable Fellow
Roundtable Fellow
Posts: 60
Joined: Wed Jul 04, 2007 9:04 am
Location: Germany

Risk/Position adjustment if the future's underlying changes?

Post by ADMP »

Hi,

the "standard" position management is to adjust the size of a new trade for the current risk (atr, volatility, etc).

What do you do on a roll date when the underlying of the new contract is "quite" different from the underlying of the old contract???
It is sometimes (not very often) the case for bond futures when the cheapest to deliver of the new contract has a maturity that is far from the one of the cheapest to deliver of the old contract. If you base your position on atr or volatility, you roll about the same amount of contracts because the atr or the volatility are not corrected for this difference of cheapest to deliver. But the 2 cheapest to deliver have different durations, and the 2 futures too, even after correcting for the different conversion factors.
Any idea or suggestion?

Alex
sluggo
Roundtable Knight
Roundtable Knight
Posts: 2987
Joined: Fri Jun 11, 2004 2:50 pm

Re: Risk/Position adjustment if the future's underlying chan

Post by sluggo »

ADMP wrote:What do you do on a roll date
Would you say more please? Do you mean
  1. What do you do on a roll date when you have a position and you don't get an exit signal?
  2. What do you do on a roll date when you have a position and you do get an exit signal?
  3. What do you do on a roll date when you have no position and you don't get an entry signal?
  4. What do you do on a roll date when you have no position and you do get an entry signal?
Or perhaps you mean something else entirely, unrelated to the items above. Would you say more?
RedRock
Roundtable Knight
Roundtable Knight
Posts: 944
Joined: Fri Jan 30, 2004 3:54 pm
Location: Arizona

Re: Risk/Position adjustment if the future's underlying chan

Post by RedRock »

If I understand you correctly, the answer would be to ignore the underlying. Presuming you have developed your method on back-adjusted data which ignored the underlying, trade it the same way. Just roll the same number of contracts. If your system does take the underlying composition or qualities of the contract in account, you would already know what to do having thought that deeply about it...

In WOTT, the author did mention Dennis et all, did evaluate the roll to contract to deem its worth in trend terms. That seemed subjective however it could be tested out.
ADMP wrote:Hi,

the "standard" position management is to adjust the size of a new trade for the current risk (atr, volatility, etc).

What do you do on a roll date when the underlying of the new contract is "quite" different from the underlying of the old contract???
It is sometimes (not very often) the case for bond futures when the cheapest to deliver of the new contract has a maturity that is far from the one of the cheapest to deliver of the old contract. If you base your position on atr or volatility, you roll about the same amount of contracts because the atr or the volatility are not corrected for this difference of cheapest to deliver. But the 2 cheapest to deliver have different durations, and the 2 futures too, even after correcting for the different conversion factors.
Any idea or suggestion?

Alex
ADMP
Roundtable Fellow
Roundtable Fellow
Posts: 60
Joined: Wed Jul 04, 2007 9:04 am
Location: Germany

Post by ADMP »

Sluggo,
sorry if I was not very explicit. I mean your first option: "you have a position and you don't get an exit signal".

RedRock,
you understood me correctly. You answer is also what I thought about this problem. It is not ideal from a money management perspective, but to change this, I will have to first make more tests as you mention (which means first to get the data on the underlying!).

Thanks.
Post Reply