"gaussian"Predict responses for new data using fitted models.
# S3 method for gaussian
predict(object, newdata, lambda.idx = NULL, Y.pred.idx = NULL,
type = "response", s = NULL, ...)Return type depends on type:
"response" or "link" (default): numeric matrix of
predicted values \(\hat{\beta}_0 + X \hat{\beta}\).
"nonzero": list of integer vectors of nonzero coefficient indices,
one element per selected lambda.
Rows correspond to observations; columns correspond to lambda.idx (or
s values when s is specified).
An object with S3 class "gaussian".
Nonempty finite numeric matrix of new observations for prediction (\(n_{new} \times d\)) with the same number of columns as the fitted design.
Positive integer indices of regularization parameters along the solution path used for prediction. By default, at most the first three fitted path points are used.
Optional row indices to subset returned predictions. NULL returns
every prediction row.
Type of prediction. "response" (default) returns predicted values;
"link" returns the linear predictor; "nonzero" returns a list
of nonzero variable indices per selected lambda.
Optional nonempty vector of finite non-negative lambda values (not
indices). When
supplied, lambda.idx is ignored. If a requested value falls between
two path lambdas the coefficients are linearly interpolated and a message is
issued. Values outside the path range are clamped to the nearest endpoint.
For type = "nonzero", the nearest fitted lambda is used instead
because support sets cannot be interpolated.
Arguments to be passed to methods.
Jason Ge, Xingguo Li, Haoming Jiang, Mengdi Wang, Tong Zhang, Han Liu and Tuo Zhao
Maintainer: Tuo Zhao <tourzhao@gatech.edu>
predict.gaussian returns predicted responses for newdata using fitted coefficients from object:
$$
\hat{Y} = \hat{\beta}_0 + X_{new} \hat{\beta}.
$$
picasso and picasso-package.