|
PureBytes Links
Trading Reference Links
|
> 1 year (Nbars = 250 or 52 respectively)
>... you then have an EQUAL basis of comparison for ALL your potential
> trades, investments, etc, and even the
In a perfect world where every stock trades in every single trading day of
the year, this would be true. Unfortunately 250 bars in one stock may
take you back 52 weeks, whereas the same number of bars may be 54 weeks or
so for another stock which may have been inactive for some of the time.
This issue is a major headache for Relative Strength Comparative(RSC)
studies, where the Index and security zero start points need to be
synchronized for a valid comparison.
More on this issue here:
http://www.metastocktools.com/URSC/URSC.htm
> Jose's Metastocktools site is great, it gave me lots of inspiration
And it gives me lots of perspiration... ;)
> (Still in Moscow, probably moving to Mumbai later this year)
Moscow to Mumbai - not far from each other on the alphabet, but quite a
change of lifestyle I'd imagine. I'd definitely leave the Cossack boots
behind. :)
jose '-)
http://www.metastocktools.com
--- In equismetastock@xxxxxxxxxxxxxxx, Tony M <t4tonym@xxx> wrote:
>
> Xuanling & other Metastock friends,
>
> Jose's Metastocktools site is great, it gave me lots of inspiration, and
I agree with Jose that the gradient depends on the X-axis timescale,
"averaging interval", or "smoothing period" that is used. However, as a
trader, investor and project manager, I have some slightly different ideas
regarding objective measurements for gradient plots. Personally, I use the
Annualized ROI% (annualized percentage Return on Investment) concept, in
which the percentage price change is converted to a common objective
measure that can then be used for validly comparing all stocks, indices,
futures, FX trades, project returns, etc.
>
> "Annualized" ROI does NOT mean that you have to take the averaging
period as a whole year, but rather that, whatever averaging period you
take, the result is then converted to what would be the return that you
would get, if the price kept changing at that same rate for a whole
year.
>
> Try this:
> MyPrice: = C; {or use EMA, or zero-phased EMA, or whatever else you
prefer}
> BarsPerYear: = 250; {use 250 here if daily, or use 52 if weekly data}
> Lag: = Nbars; {...use whatever you want here for your averaging
interval.
> Can be equal to a year, in which case use 250 for
daily or 52 for weekly data,
> OR it can be as short as 1 (bar) , if you want to
take "instantaneous" slope}.
>
> AnnualizedROIgradient:= 100 * (MyPrice/Ref(MyPrice,-Lag) -1) *
BarsPerYear / Lag
>
> This allows you to select whatever Lag (smoothing) interval you want,
although the obvious candidates for stocks are:
> 1 day (Nbars = 1 bar on daily charts. This is very noisy, but good to
show extreme values),
> 1 week (Nbars = 5 bars on daily charts, or 1 bar on weekly charts),
> 1 month (Nbars = 21 or 4 respectively)
> 1 quarter (Nbars = 63 or 13 respectively)
> 1 year (Nbars = 250 or 52 respectively)
>
> As Jose says, of course the results will differ depending on the
timescale you look at, but that's true for most other indicators, and what
you use really depends on the timescale that you are concerned about. The
advantage of the annualized ROI concept for gradient is that, once you
decide on the interval that you need to use, which should be consistent
with your trading or investment timeframe, you then have an EQUAL basis of
comparison for ALL your potential trades, investments, etc, and even the
interest (if any) that is paid on your bank account. :-)
>
> Cheers,
> from Tony
> (Still in Moscow, probably moving to Mumbai later this year)
>
>
> ====================
> ----- Original Message ----
> From: Jose Silva <josesilva22@xxx>
> To: equismetastock@xxxxxxxxxxxxxxx
> Sent: Saturday, April 7, 2007 2:46:31 PM
> Subject: [EquisMetaStock Group] Re: Find gradient of MA
>
> Xuanling, the gradient of any plot is always subjective - it depends on
> the x-axis' (time/date) settings. Apart from 0 degrees (flat), it is
> not possible to measure any plot gradient objectively.
>
> Take a look at these indicators below from
> http://www.metastocktools.com/#metastock :
>
> EMA-slope - EMA synthetic slope indicator.
>
> LinReg formulae - Linear Regression Indicator/Slope formulae.
>
> LinReg Slope - Linear Regression synthetic slope, displays output in
> 0~100% or 0~90 degrees.
>
>
> jose '-)
> http://www.metastocktools.com
>
>
>
> --- In equismetastock@ yahoogroups. com, "CXL2" <cxl2@> wrote:
>
> Hi,
>
> I will like to compute the gradient of Moving Average (or data array)
> for a given period. Based on the gradient, I can know whether the
> moving average is how steep or flat it is and its direction within a
> given period. Is there anyway of doing so?
>
> Thanks alot!
>
> Xuanling
Yahoo! Groups Links
<*> To visit your group on the web, go to:
http://groups.yahoo.com/group/equismetastock/
<*> Your email settings:
Individual Email | Traditional
<*> To change settings online go to:
http://groups.yahoo.com/group/equismetastock/join
(Yahoo! ID required)
<*> To change settings via email:
mailto:equismetastock-digest@xxxxxxxxxxxxxxx
mailto:equismetastock-fullfeatured@xxxxxxxxxxxxxxx
<*> To unsubscribe from this group, send an email to:
equismetastock-unsubscribe@xxxxxxxxxxxxxxx
<*> Your use of Yahoo! Groups is subject to:
http://docs.yahoo.com/info/terms/
|