qtlmt (version 0.1-1)

sureStep: SURE model selection

Description

Select a SURE model via model selection.

Usage

sureAdd1(obj, y, x, range=NULL, iter=250, tol=1e-8, ext=FALSE)

sureDrop1(obj, y, x, range=NULL, iter=250, tol=1e-8, ext=FALSE)

sureStep(y, x, v, lower, upper, k, direction=c("both","backward",
   "forward"), iter=250, max.terms=200, steps=1000, tol=1e-8)

sureStep2(obj, y, x, cv, direction=c("both","backward","forward"),
   range=NULL, iter=250, steps=1000, tol=1e-8, ext=FALSE)

Arguments

obj

initial model in model search; can be an object of sureEst.

y

n by p matrix, whose columns are dependent variables.

x

n by m matrix, whose columns are predictor variables to select from.

cv

critical value used in the likelihood ratio test to determine adding/dropping a term.

v

list; v[[j]] indicates which x's to start with for y[,j].

range

list; range[[j]] indicates which x's (all if NULL) correspond to which y[,j].

lower

list in the format of v; lower scope of the model.

upper

list in the format of v; upper scope of the model.

k

penalty, 0 if missing or

direction

forward selection, backward elimination or stepwise.

iter

maximum number of iterations in a numerical process to estimate model parameters.

max.terms

maximum number of terms in the final model.

steps

maximum number of search steps.

tol

convergence tolerance.

ext

whether to perform an extensive search.

Value

• a list with the following components:
• loglik:
  log-likelihood of the model
• b:
  estimates of model coefficients
• sigma:
  estimates of residual variance-covariance
• v:
  variables selected in the model

Examples

data(etrait)
x<- as.matrix(3/2-mdat)
y<- as.matrix(traits)[,1:3]
v<- list()
upper<- list()
for(k in 1:ncol(y)){
   v[[k]]<- numeric(0)
   upper[[k]]<- 1:ncol(x)
}
g1<- sureStep(y, x, v=v, upper=upper, k=19)
g2<- sureStep2(g1, y, x, cv=1/2, ext=FALSE)