loess() is a 'local' fitting procedure, which means that it fits a
different regression at each prediction point. (In principle, for there
are computational short cuts: see ?loess.control.) Since each fit is only
used for a single prediction, its coeffiicients are not of interest and
not made
May I inquire about how to obtain the coefficients from a loess fit of
multiple covariates? I have tried terms and coef, but it does not provide
me with the coefficients to identify the terms of the entire regression
equation.
Many thanks
Paul
--
Paul R. Hutson, Pharm.D.
2 matches
Mail list logo
