|
Hi Dan,
Sorry for the blind ally with the Hull MA -- I have substituted it in a
number of other formula for a MA or EMA with nice results (much smoother and
responsive curving). I hadn't thought thru the consequence of imaginary
numbers -- but then sometimes our profits seem all too imaginary, so I should
have. <VBG>
I'll have to take a look at your EMA version some more and play with
it some. Thanks again for the insight.
Peace and Justice --- Patrick
----- Original Message -----
Sent: Tuesday, May 23, 2006 8:57 PM
Subject: Re: [amibroker] Exponential
Standard Deviation (EStDev)
Thanks for the pointer to the Hull MA, Patrick. I found a
definition and
experimented with it, substituting HMA for EMA in my
formula (based on
"sample stdev") and also the one that Paul Ho provided
(based on "population
stdev").
HMA is apparently too fast
for standard deviation calculations because the
result is sometimes
imaginary, due to taking the square root of a negative
number.
I'm
not sure why using EMA works out, and maybe it does fail in some
situations. Even DEMA is too fast. I suppose we could always
just use Nz()
to constrain it to zero, but who knows what the statistical
meaning is
anymore. For that matter, why should we assume that
standard deviation means
anything regarding the non-normal misbehaving
stock market? Oh well,
whatever works :)
dan
On
Sunday 21 May 2006 11:52 pm, NW Trader wrote:
> Hi Dan,
>
>
Try using a Hull MA -- it's a lot faster.
>
> Peace and
Justice --- Patrick
Please note that this group is for discussion between users only.
To get support from AmiBroker please send an e-mail directly to
SUPPORT {at} amibroker.com
For other support material please check also:
http://www.amibroker.com/support.html
SPONSORED LINKS
YAHOO! GROUPS LINKS
|