|
PureBytes Links
Trading Reference Links
|
Hi Roy,
Basically, Linear Regression (LR) is a trend line drawn between two
points using the least squares fit method, which includes all data in
between.
Linear Regression Slope (LRS) is the constant slope of the LR,
measured in a basic LR-Ref(LR,-1) way:
positive LRS values = LR uptrending;
negative LRS values = LR downtrending.
The current LRS value is plotted on the last LR bar.
Linear Regression Indicator (LRI) is the end point chart value of the
LR line {LastValue(LR)}. Plot a 21-bar Linear Regression trendline,
as well as a 21-period LRI, move the LR line around the chart, and
watch the LR's last point track the LRI exactly across the chart.
Exponential Regression, Polynomial Regression, and others, are way
beyond my basic understanding of this useful tool.
And then there is Past Life Regression... :)
jose '-)
--- In equismetastock@xxxxxxxxxxxxxxx, "Roy Larsen" <rlarsen@xxxx>
wrote:
> Hi Jose
>
>
> > Roy, both posted Linear Regression Slope formulae outputs are so
close
> > to the MetaStock canned LRS version, that in my view it isn't
worth
> > the extra code complexity and processing overheads.
>
> Interesting. I need to think about the signifcance of this. My
thinking is that the linear
> regression of a line is different from the line itself, and
therefore the slope of a line should be
> different from the slope of the linear regression of that line. Have
I got it wrong? Perhaps I
> should do some of my own testing rather than go with unfounded
assumptions.
>
> Roy
>
> > ---8<------------------
> > pds:=Input("Periods",5,50,10);
> > LinRegSlope(P,pds)
> > ---8<------------------
> >
> > ---8<------------------
> > A:=Tema(C,75);
> > pds:=10;
> > LinRegSlope(A,pds)
> > ---8<------------------
> >
> > jose '-)
> >
> >
> > --- In equismetastock@xxxxxxxxxxxxxxx, "Roy Larsen" <rlarsen@xxxx>
> > wrote:
> > > Bill
> > >
> > > I have this formula for "slope of line".
> > >
> > > {Slope of Line}
> > > Periods:=Input("Periods",5,50,10);
> > > Slope:=((Sum(Cum(1)*P,Periods))-
> > > (Sum(Cum(1),Periods)*Sum(P,Periods)/Periods))/
> > > ((Sum(Power(Cum(1),2),Periods))-
> > > (Power(Sum(Cum(1),Periods),2)/Periods));
> > > Slope;
> > >
> > > Here's virtually the same thing applied to a 75 period Tema of
> > CLOSE.
> > >
> > > {Slope of Tema}
> > > A:=Tema(C,75);
> > > Pds:=10;
> > > Slope:=((Sum(Cum(1)*A,Pds))-
> > > (Sum(Cum(1),Pds)*Sum(A,Pds)/Pds))/
> > > ((Sum(Power(Cum(1),2),Pds))-
> > > (Power(Sum(Cum(1),Pds),2)/Pds));
> > > Slope;
> > >
> > > I'm unaware or who the author is but think I picked these up
from
> > the Guppy site ages ago.
> > >
> > > Roy
------------------------ Yahoo! Groups Sponsor ---------------------~-->
Buy Ink Cartridges or Refill Kits for your HP, Epson, Canon or Lexmark
Printer at MyInks.com. Free s/h on orders $50 or more to the US & Canada.
http://www.c1tracking.com/l.asp?cid=5511
http://us.click.yahoo.com/mOAaAA/3exGAA/qnsNAA/BefplB/TM
---------------------------------------------------------------------~->
Yahoo! Groups Links
<*> To visit your group on the web, go to:
http://groups.yahoo.com/group/equismetastock/
<*> 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/
|