EstAllModels: Estimate all submodels

In sebastianankargren/FMA: Frequentist Model Averaging

Description

This function estimates all submodels.

Usage

EstAllModels(X, Xnew, y, s)

Arguments

X	A design matrix.
Xnew	Design matrix used for predictions.
y	Vector of the response.
s	Matrix where the rows correspond to a submodel.

Value

Returns a list with the following:

SEs	Variances of the submodels.
coefficients	Coefficients from each submodel.
K	The number of included variables in each submodel.
etilde	Matrix of residuals (for MMA). Each column corresponds to residuals from a submodel.
eJMA	The matrix of error terms for JMA. Each column corresponds to jackknife residuals in the submodels.
AIC	The AIC values for the models.
BIC	The BIC values for the models.
preds	The predictions from all submodels.

See Also

Generate.S.

Examples

## Generate some data
Xdata <- matrix(rnorm(100), 25, 4)
ydata <- matrix(rowSums(Xdata) + rnorm(25), 25, 1)
Xnewdata <- matrix(rnorm(100), 25, 4)

## Get the submodel matrix
smatrix <- Generate.S(Fix.X = 1, Potential.X = 3, avertype = "nested")

## Estimate all submodels
allMods <- EstAllModels(X = Xdata, Xnew = Xnewdata, y = ydata, s = smatrix)

sebastianankargren/FMA documentation built on May 29, 2019, 4:57 p.m.