rlmFit {LVSmiRNA}R Documentation

Fitter Functions for Robust Linear Models

Description

These are the basic computing engines called by RLM used to fit robust linear models. These should not be used directly unless by experienced users.

Usage

rlmFit(x, y, maxit=20L, k=1.345, offset=NULL,method=c("joint","rlm"),
cov.formula=c("weighted","asymptotic"),start=NULL, error.limit=0.01)

Arguments

x

design matrix of dimension n * p.

y

vector of observations of length n, or a matrix with n rows.

maxit

the limit on the number of IWLS iterations.

k

tuning constant used for Huber proposal 2 scale estimation.

offset

numeric of length n. This can be used to specify an a priori known component to be included in the linear predictor during fitting.

method

currently, only method="rlm.fit" is supported.

cov.formula

are the methods to compute covariance matrix, currently either weighted or asymptotic.

start

vector containing starting values for the paramter estimates.

error.limit

the convergence criteria during iterative estimation.

Value

a list with components

coeffecients

p vector

Std.Error

p vector

t.value

p vector

cov.matrix

matrix of dimension p*p

res.SD

value of residual standard deviation

...

Author(s)

Stefano Calza <stefano.calza@biostatistics.it>, Suo Chen and Yudi Pawitan.

References

Yudi Pawitan: In All Likelihood: Statistical modeling and inference using likelihood. Oxford University Press. 2001.

See Also

RLM which you should use for robust linear regression usually.

Examples


set.seed(133)
n <- 9 
p <- 3
X <- matrix(rnorm(n * p), n,p) #no intercept
y <- rnorm(n)

RLM.fit <- rlmFit (x=X, y=y)

[Package LVSmiRNA version 1.32.0 Index]