sn.Einfo {sn} | R Documentation |
Computes expected Fisher information for parameters of simple sample having one-dimensional skew-normal (SN) distribution or regression model having SN errors, in the DP and CP parametrizations.
sn.Einfo(dp=NULL, cp=NULL, n=1, x=NULL)
dp, cp |
vector of direct or centred parameters, respectively; one of the two vectors must be supplied, but not both. See below for more details. |
n |
sample size; if this parameter is supplied, then x must
not be.
|
x |
design matrix of the regression model; if this parameter is
supplied, then n must not be.
|
a list containing the following components:
dp, cp |
DP and CP parameters; one of the two vectors is the one supplied on input, the other one matches the previous one in the alternative parametrization |
info.dp, info.cp |
matrices of Fisher expected information in the two parametrizations |
se.dp, se.cp |
vectors of standard errors in the two parametrizations |
D |
matrix of derivatives of DP parameters with respect to CP parameters |
When x
is not set, then a simple random sample is assumed and a
matrix x
with a single column of all 1's is constructed; in this
case, the supplied vector dp
or cp
must have length 3.
If x
is set, then the supplied vector of parameters must have
length ncol(x)+2
.
For a description of the DP and CP parametrizations, see the reference below.
Azzalini, A. and Capitanio, A. (1999). Statistical applications of the multivariate skew-normal distribution. J.Roy.Statist.Soc. B 61, 579602.
info <- sn.Einfo(dp=c(0,1,5), n=3) # data(ais) M <- model.matrix(~ais$"Ht") mle <- sn.mle(X=M, y=ais$"Wt", plot.it=FALSE) info <- sn.Einfo(cp=mle$cp, x=M)