|
PureBytes Links
Trading Reference Links
|
> This is trickier than it seems.
> But that is not the case since the high of data1 and the low of
> data2 do not occur at the same time within the bar.
That's right. (Of course it's right, Bob posted it!! :-)
The only truly correct way to calculate the OHLC of a spread is
to monitor both data series on a tick-by-tick basis. Look at the
spread value after each tick, and record the highest & lowest
values of the spread.
Even that is approximate, since the two data series probably
don't update at the same instant. E.g. say you've got a very
active issue in data1, like ES, and a less active issue in data2,
like SP. When ES updates, you have the current value for ES.
But the last tick for SP might be many minutes old. Unless the
ticks happen at roughly the same time, the spread calculation
isn't valid.
For a case like ES/SP, the best way to do it would be to
calculate the spread value when *SP* updates. You can assume
that ES has updated within a few seconds or less, so your spread
calculation should be pretty accurate.
But what if you have two slow issues, like the next-out contracts
of SP and ND? There might not be any time when the two of them
update at nearly the same time. So you can never calculate a
truly accurate spread.
Gary
|