Make predictions from a fitted "abess" object.

# S3 method for abess
predict(object, newx, type = c("link", "response"), support.size = NULL, ...)

Arguments

object

An "abess" project.

newx

New data used for prediction. If omitted, the fitted linear predictors are used.

type

type = "link" gives the linear predictors for "binomial", "poisson" or "cox" models; for "gaussian" models it gives the fitted values. type = "response" gives the fitted probabilities for "binomial" and "ordinal", fitted mean for "poisson" and the fitted relative-risk for "cox"; for "gaussian", type = "response" is equivalent to type = "link".

support.size

An integer value specifies the model size fitted at given support.size. If support.size = NULL, then the model with the best tuning value would be returned. Default: support.size = NULL.

...

Additional arguments affecting the predictions produced.

Value

The object returned depends on the types of family.