rlmFit {LVSmiRNA} | R Documentation |
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.
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)
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. |
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 |
...
Stefano Calza <stefano.calza@biostatistics.it>, Suo Chen and Yudi Pawitan.
Yudi Pawitan: In All Likelihood: Statistical modeling and inference using likelihood. Oxford University Press. 2001.
RLM
which you should use for robust linear regression usually.
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)