[Date Prev][Date Next][Thread Prev][Thread Next][Date index][Thread index]
st: RE: Fixed intercept in linear regression, How?
> I'd like to do a linear regression with the intercept fixed
> at 1, i.e.,
> y = 1 + bx
> I found it's quite simple in SAS and splus to do so, but I
> was not able
> to figure out how to fullfill this in stata.
Others have suggested the trick of subtracting 1
from the response and using -regress, noconstant-.
For completeness, I will mention the use of -constraint-
followed by -cnsreg-.
The two are not quite equivalent. For example, -cnsreg-
refuses to print R-squared, which is possibly the better
response in this circumstance.
* For searches and help try: