predict.mave {MAVE}

Make predictions based on the dimension reduction space

Description

This method make predictions based the reduced dimension of data using mars function.

Usage

## S3 method for class 'mave'
predict(object, newx, dim, ...)

## S3 method for class 'mave.dim'
predict(object, newx, dim = "dim.min", ...)

Arguments

object

the object of class 'mave'

newx

Matrix of the new data to be predicted

dim

the dimension of central space or central mean space. The matrix of the original data will be multiplied by the matrix of dimension reduction directions of given dimension. Then the prediction will be made based on the data of given dimensions. The value of dim should be given when the class of the argument dr is mave. When the class of the argument dr is mave.dim and dim is not given, the function will return the basis matrix of CS or CMS of dimension selected by mave.dim

...

further arguments passed to mars function such as degree.

Value

the prediced response of the new data

See Also

mave for computing the dimension reduction space and mave.dim for estimating the dimension of the dimension reduction space

Examples

X = matrix(rnorm(10000),1000,10)
beta1 = as.matrix(c(1,1,1,1,0,0,0,0,0,0))
beta2 = as.matrix(c(0,0,0,1,1,1,1,1,0,0))
err = as.matrix(rnorm(1000))
Y = X%*%beta1+X%*%beta2+err

train = sample(1:1000)[1:500]
x.train = X[train,]
y.train = as.matrix(Y[train])
x.test = X[-train,]
y.test = as.matrix(Y[-train])

dr = mave(y.train~x.train, method = 'meanopg')

yp = predict(dr,x.test,dim=3,degree=2)
#mean error
mean((yp-y.test)^2)

dr.dim = mave.dim(dr)

yp = predict(dr.dim,x.test,degree=2)
#mean error
mean((yp-y.test)^2)