Methods_LASSO: LASSO methods

Description

Predicted values for a provided matrix of predictors X

Usage

# S3 method for LASSO
coef(object, ..., i=NULL)
# S3 method for LASSO
fitted(object, ...)

Value

Method coef returns a matrix that contains, for each value of lambda (in columns), the predicted values corresponding to each row of the matrix X.

Method fitted returns fitted values Xβ

Arguments

object: An object of the class 'LASSO' returned either by the function 'LARS' or 'solveEN'
...: Other arguments: X (numeric matrix) scores for as many predictors there are in ncol(object$beta) (in columns) for a desired number n of observations (in rows)
i: (integer vector) Index columns of matrix 'Gamma' to be considered

Author

Marco Lopez-Cruz (maraloc@gmail.com) and Gustavo de los Campos

Examples

Run this code

  require(SFSI)
  data(wheatHTP)
  
  y = as.vector(Y[,"E1"])  # Response variable
  X = scale(X_E1)          # Predictors
  
  # Training and testing sets
  tst = which(Y$trial %in% 1:10)
  trn = seq_along(y)[-tst]

  # Calculate covariances in training set
  XtX = var(X[trn,])
  Xty = cov(X[trn,],y[trn])
  
  # Run the penalized regression
  fm = solveEN(XtX,Xty,alpha=0.5)   
  
  # Regression coefficients
  B = coef(fm)
  
  # Predicted values
  yHat1 = fitted(fm, X=X[trn,])  # training data
  yHat2 = fitted(fm, X=X[tst,])  # testing data
  
  # Penalization vs correlation
  plot(-log(fm$lambda[-1]),cor(y[trn],yHat1[,-1]), main="training")
  plot(-log(fm$lambda[-1]),cor(y[tst],yHat2[,-1]), main="testing")