InvestorsHub Logo

Rien

05/07/02 8:38 AM

#460 RE: stiv #459

Hi stiv, XDEV in the version as you used it needs constant tuning. That is probably the reason for it's "mediocre" performance. If you would optimize XDEV for each 100 day cycle in that 5 year period, it should do much better.

About the "all or nothing" approach, I am not sure. I have played with the idea myself, but have come to no real conclusion. I guess I need to do a simulation on that myself to find out just what the advances and drawbacks might be.


Best,
Rien.

Myst

05/07/02 5:18 PM

#463 RE: stiv #459

Stiv!

Interesting werk! I'd really like to know what kind of formula's you've worked up for the stops. Can you send me the sheet?

100% in and out sounds like a whole nuther ball of wax, but hey.....if it works than it works.

I've been trying to make X_DEV more and more self-adjusting. The default method is a step in the right direction, because at least the deviation bands are self-adjusting but......(there's always a damn "but").....I've found that basing them off a 12 day sma is great for some stocks but not so great for other stocks. Changing the sma in the default mode makes the bands wider or narrower, which again is ok for some stocks and not for others. And there still is the problem with longer term studies. With your help from a previous email you sent me, (Stiv is an excel guru, btw) I finally figured out the excel "max" feature, which was a great help utilizing Williams%R. I now have an excel formula for williams%r if anyone wants it. Pretty simple actually.

The problem I currently face with the default method....besides the 12 sma problem....is that it will sell off shares to quickly, or burn cash too quickly in strong trends. Works great in range bound moves, but there seems to be more need for a balance. AIM uses a ratio between Stock Value and Portfolio Balance to balance out buys and sells. I've fiddled some with that idea, but came away unsatisfied.

Perhaps your use of stops may be of help.

Thanks for your continued efforts. Your previous spreadsheets have been a great help and are very much appreciated!



~Myst~