Re: multiple linear regression

snip
I've looked round on the web and from what I can find it seems
everybody uses dummy variables of 0 and 1. Why is this? If centering
is important to interpreting the interaction term, why don't people
use dummy variables of -1 and 1?

Multiplying by 1 turns on an effect. Multiplying by zero turns it off.
If D equals 1 or 0, then the equation

y =a+ bX + cD + dDX+ u

equals a+ bX when D=0 and
(a+c) + (b+d)X

when D=1

-*** Startz
.

Quantcast