predict.cggd {cggd} | R Documentation |
While cggd
produces the entire path of solutions, predict.cggd
allows one to extract a prediction at a particular point along the path.
## S3 method for class 'cggd': predict(object, newx, t, type = c("fit", "coefficients"), mode=c("k","t"), ...)
object |
A fitted cggd object |
newx |
If type="fit", then newx should be the x values at which the fit is required. If type="coefficients", then newx can be omitted. |
t |
A value specifying the desired model. Its values depends on the mode argument. By default (mode="k"), t should take on values between 0 and max k (e.g., a step of 1.3 means .3 of the way between step 1 and 2.) |
type |
If type="fit", predict returns the fitted value. If type="coefficients", predict returns the coefficients. |
mode |
Mode="k" means the t= argument indexes the cggd step number, and the coefficients will be returned corresponding to the values at step t. If mode="t", then t is a time on the coefficient path. |
... |
additional parameters. |
Either the fitted value, or a vector of coefficients.
Cun-Hui Zhang and Ofer Melnik
Cun-Hui Zhang (2007) "Continuous Generalized Gradient Descent" Journal of Computational and Graphical Statistics ; see also http://stat.rutgers.edu/~cunhui/software/CGGD.html.
plot, cggd, cv.cggd
data(housing) attach(housing) object <- cggd(x,y) coef4.1 <- predict(object, t=4.1,type="coefficients", mode="k") detach(housing)