[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

[amibroker] Re: Exploration for tight rising channels



PureBytes Links

Trading Reference Links

Hi Graham,

Thanks for responding. My interpretation of your response is that it 
identifies the first and last values of the efficiency ratio's (ER's) 
denominator (the volatility part). What I am actually looking for, 
and probably didn't explain very well, are the first and last values 
of the numerator (the direction part). If I am misinterpreting your 
response, please forgive my ignorance.

For my adaptation of the ER, I am looking for the y values of the 
first and last points of a static (non-moving/straight) regression 
line instead of the closing price at the beginning and the end of the 
period.

Here's what I have started with ...

Filter = 1;

Periods = BarIndex() - BeginValue (BarIndex());
/* sets the start of the period with the BeginValue marker on the 
chart and the end of period as the end of the chart (i think) */

Direction = LinearReg(Close,Periods) - BeginValue(LinearReg
(Close,Periods));
/* I want to substitute the end and beginning points on a static 
linear regression line for the standard definition of Direction which 
is the difference between the closing prices at the beginning and end 
of the period = C - Ref(C, -periods). My code obviously does not 
accomplish what I want because it returns the difference between the 
values of 2 end points on a moving regression line */

Volatility = Sum(abs(C-LinearReg(C,Periods)), periods);
*/ What I want to do here is calculate the absolute value of each 
closing price from each point on the static linear regression line 
and sum the total as a substitute for the standard definition of 
Volatility which = Sum(Abs(ROC(C,1), periods) */

ER = Direction / Volatility;

Maybe this explanation will better illustrate what I would like to do.

Tim

--- In amibroker@xxxxxxxxxxxxxxx, Graham <kavemanperth@xxx> wrote:
>
> You can use lastvalue if you are testing to the last bar of the 
chart,
> otherwise if you want to do this over a range within the chart 
history
> you can use status to find the first and last values for that range
> for use in analysis window it will take the date range as used. I 
have
> also added lastvalue so that it will always give these values even
> during the range
> 
> first = lastvalue( valuewhen( status("firstbarinrange"), Sum(Abs(C -
> LinearReg(C, periods)), periods) ) );
> last = lastvalue( valuewhen( status("lastbarinrange"), Sum(Abs(C -
> LinearReg(C, periods)), periods) ) );
> 
> if you want this in a chart then you can use the range markers 
(mouse
> double click gives the lines with triangles at bottom) and 
beginvalue
> and endvalue functions
> 
> 
> --
> Cheers
> Graham
> AB-Write >< Professional AFL Writing Service
> Yes, I write AFL code to your requirements
> http://e-wire.net.au/~eb_kavan/ab_write.htm
> 
> 
> 
> On 4/15/06, timgadd <timgadd@xxx> wrote:
> > I'd like to employ a variation of Kaufman's Efficiency Ratio to 
help
> > screen for stocks in tight rising channels. Specifically, I'd 
like to
> > compare a stock's ROC weighted by it's Efficiency Ratio to that 
of a
> > market index to screen out stocks that are outperforming the 
index on
> > a volatiliy adjusted basis. (I realize there are other ways to
> > approach this, but I'd like to try this approach).
> >
> > The standard version of the Efficiency Ratio is defined as the
> > absolute value of direction over volatility where ...
> >
> >     Direction = C - Ref(C, -periods);
> >     Volatility = Sum(Abs(ROC(C,1), periods);
> >
> > I'd like, instead, to define Direction as the last (y) value of a
> > static (non-moving) linear regression line minus the first (y) 
value
> > of the static linear regression line.
> >
> > I'd like to define Volatility as the "scatter" at each point along
> > the static linear regression line ... something like ...
> >
> >     Sum(Abs(C - LinearReg(C, periods)), periods);
> >
> > ... except that this uses the moving linear regression 
calculation.
> > I'd like to keep the start point of the regression line constant 
so
> > that I can measure the efficiency ratio from a static reference 
point
> > (a technical low).
> >
> > I've stumbled through some code for a static linear regression 
line
> > (link below), but cannot figure our how to determine the beginning
> > and ending points.
> >
> > http://www.amibroker.com/library/detail.php?
id=193&hilite=LASTVALUE
> >
> > Any suggestions?
> >
> >
> >
> >
> >
> > 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
> >
> >
> > Yahoo! Groups Links
> >
> >
> >
> >
> >
> >
> >
> >
>






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

 
Yahoo! Groups Links

<*> To visit your group on the web, go to:
    http://groups.yahoo.com/group/amibroker/

<*> To unsubscribe from this group, send an email to:
    amibroker-unsubscribe@xxxxxxxxxxxxxxx

<*> Your use of Yahoo! Groups is subject to:
    http://docs.yahoo.com/info/terms/