SVAPO possible?

[jonesenberg]

After a number of searches here and elsewhere, I've not been able to find any hints that the promising SVAPO indicator (as mentioned frequently in S&C) is available or even possible for NT. Any thoughts?

[NinjaTrader_Josh]

The difficulty with SVAPO is that in the literature it is defined with using a TEMA of period 1.6. Having a period of 1.6 makes no sense in the context of NinjaTrader or even in the context of TEMA. If you decide to disregard that then SVAPO is possible.

[jonesenberg]

According to an email alert, thrunner made an interesting reply to this thread but it's not showing. He also attached 2 files and some code, but none of it is posted.

thrunner: The EMAD indicator you speak of.. is this an interpretation of the entire SVAPO indicator or is it a piece of it? Thanks

[thrunner]

According to an email alert, thrunner made an interesting reply to this thread but it's not showing. He also attached 2 files and some code, but none of it is posted.
thrunner: The EMAD indicator you speak of.. is this an interpretation of the entire SVAPO indicator or is it a piece of it? ThanksI am sorry, but I deleted the code once I realized it didn't work as intended. NT does not allow non integer (eg double) for dataseries periods (lengths).

The other languages such as EL, MS, AFL do allow non integers, that is why their EMA and TEMA could have decimal inputs.

The other point I was trying to make was that EMA is just an approximation, so that using decimal periods is not completely nonsensical.

Here is the EMA from EL, there were two types of approximation, one the original SmoothingFactor and the current.

{ Exponential average }
inputs:
Price( numericseries ),
Length( numericsimple ) ; { this input assumed to be a constant >= 1 }
variables:
SmoothingFactor( 2 / ( Length + 1 ) ) ; //
//original version SmoothingFactor( 1 / Length ) ;
if CurrentBar = 1 then
XAverage = Price
else
XAverage = XAverage[1] + SmoothingFactor * ( Price - XAverage[1] ) ;


SVAPO could be made to work in NT, you just have to approximate TEMA; I am sure Josh and others already have a close approximation of SVAPO in NT.

[NinjaTrader_Josh]

Basically it is possible, but it would require tweaking.

[jonesenberg]

Sorry to revive such an old thread, but I came back to it and can't help but wonder how close an approximation we can get to SVAPO by simply rounding the TEMA used up to 2.0?

[NinjaTrader_Josh]

Interesting question that I am unqualified to answer, but I would imagine it would have large implications depending on how much rounding is done.