Re: "Regression" average VS the "Arithematic" Average
- From: "Nick" <tulse04-news1@xxxxxxxxxxx>
- Date: Mon, 19 Feb 2007 10:19:59 -0000
"Brablo" <gestureofrespect@xxxxxxxxx> wrote in message
news:1171855401.344234.270730@xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx
Here is the data (all ~487 or so points of the S&P 500).
You totally misunderstood what I'm saying. I did *NOT* set up one
experiment with a constant. Let me re-clarify what I did.
I didn't say that you did.
By your own statement you said that:
"However, when I attempt to regress the market cap to the book value
without using a constant, I get something totally different!"
ie when you did calculated the relationship using the regression you assumed
a constant of zero.
This will give you a different slope than if you calculate the constant
using the data.
If the intercept with the y-axis (the constant) is greater than zero, then
forcing a constant of zero will increase the slope of the line.
If the constant is negative, then forcing a constant of zero will reduce the
slope of the line.
In both cases, the line will pass through or near the point (E(x), E(y))
where E(x) is the mean of x.
Nick
.
- Follow-Ups:
- References:
- Prev by Date: Re: Disjoint union: a joke, or what?
- Next by Date: Explicit Solutions for the Euler Top (torque free)
- Previous by thread: Re: "Regression" average VS the "Arithematic" Average
- Next by thread: Re: "Regression" average VS the "Arithematic" Average
- Index(es):
Relevant Pages
|
