Richard, Brooke (as keeper of the formulae), et al..... Moany of the old guard remember Jack Landis' weighted Stochastics formula that he presented in Biloxi.......I wasn't there yet if he re-presented it in New Jersey, but it is (was?):
Fml ("Landis") =
((Stoch(8,3)*.05)+(Stoch(89,21)*.43)+(Stoch(55,13)*.26)+(Stoch(34,8)*.16))
Many of us have tried to use Stoch (55,21) as a surrogate with only mixed results. I have Jack's formula on my default screen and wouldn't look at a stock without first knowing what that indicator was doing.......I now seem to remember that Brooke posted a piece about using either Jack's formula OR Stoch (55,21) with a 5 period simple moving average as the trigger.
Anyway......I've been playing around with Slopes, and tried to put one on Jack's formula. That worked out OK in MSWIN, but I later found that a simple Rate-of Change formula in MetaStock produced the same graph, but in a much more workable set of parameters. So I created the following:
Fml ("LandSlide") = ROC(Fml("Landis"),2,%)
Looks pretty good for both short and intermediate entries when put in as a Histogram with a Zero cross trigger.
But then I went further.....
Fml ("LandRush") = ROC(ROC(Fml("Landis"),2,%),2,$)
This formula is more trigger-happy, but it seems to signal a lot of short term moves, and also might be used for exits.
Any thoughts or ideas??
BW Chan |
