|
PureBytes Links
Trading Reference Links
|
Hi Jim,
I never applied the adaptive system to stochastics although it can be done.
The reason I am not inclined to do that is because we use these oscillators
to identify momentum or overbought/oversold on a short term basis and while
SMOOTHING them makes them less subject to whipsaws, making them ADAPTIVE will
take from them their very usefulness since they may not be short term
anymore. Remember the THEORETICAL benefit of changing the smoothing constant
in the ama is to change the length of the smoothing constant(number of days
in the m/a) depending on if the market is trending or not. Otherwise it has
no benefit over non-adaptive m/a's. Exponential smoothing does not change
the # of periods lookback of an indicator or m/a but adaptivity does, you
could have a 9 day m/a now and maybe a two day m/a elsewhere on the chart on
your ama.
Incidentally while KAUFMAN suggests measuring the ratio of noise to price
movement to adjust the ama, Chande recommends using a momentum oscillator's
readings to adjust the term of the ama. Similar results though. You need to
go much further than either to get a good trading system, IMHO.
Best wishes,
James in New Jersey
>
>
> James,
> Have you done any work on adaptive oscilators? Such as an andaptive
> stochastics? Would you use the same type of formula to do that with
> stochastics? Thanks.
> --Jim
Attachment Converted: "c:\eudora\attach\vcard177.vcf"
|