predict.cv.grpreg: Model predictions for a fitted 'cv.grpreg' object
In debinqiu/grpss: Group Screening and Selection

Description Usage Arguments Details Value Author(s) See Also Examples

Similar to usual predict methods and predict.cv.grpreg in grpreg package.

1
2
3

## S3 method for class 'cv.grpreg'
predict(object, newdata, type = c("response", "class",
  "probability"), ...)

`object`	A fitted "`grpreg`" object from `grpss`, or `cv.grpreg` function.
`newdata`	Optionally, a matrix or data frame where to predict. If omits, the fitted predictors are used.
`type`	The type of prediction: "`response`" gives the fitted values; `"class"` returns the predicted class for the binomial outcome; "`probability`" returns the predicted probabilities for the logistic regression.
`...`	Not used.

This function gives the predictions at newdata or all predictors if the argument newdata is not supplied. Typically, type = "response" is used for linear or poisson regression, and type = "class" or type = "probability" is used for logistic regression.

The predicted values depending on the type.

Debin Qiu, Jeongyoun Ahn

grpss

library(MASS)
set.seed(23)
n <- 30 # sample size
p <- 3  # number of predictors in each group
J <- 50  # group size
group <- rep(1:J,each = 3)  # group indices
X <- mvrnorm(n,seq(0,5,length.out = p*J),diag(p*J))
beta <- runif(12,-2,5)
mu <- X%*%matrix(c(beta,rep(0,p*J-12)),ncol = 1)

# linear regression with family = "gaussian"
y <-  mu + rnorm(n)

## with cross-validation
gss11 <- grpss(X,y,group,select = TRUE,cross.validation = TRUE)
predict(gss11)  # fitted values
predict(gss11, newdata = X[1,])  # predicted values at given 'newdata'

# logistic regression with family = "binomial"
set.seed(23)
y1 <- rbinom(n,1,1/(1 + exp(-mu)))
gss21 <- grpss(X,y1,group, criterion = "gDC",select = TRUE,
               family = "binomial")
predict(gss21)