sn (version 0.33)

msn.cond.plot: Plot of the density of a conditional skew-normal variate

Description

Plot of the exact and of the approximate density function of a multivariate skew-normal variate conditionally on the values taken on by some components.

Usage

msn.cond.plot(xi, Omega, alpha, fixed.comp, fixed.values, n=35)

Arguments

xi

a numeric vector of length k, say, giving the location parameter.

Omega

a covariance matrix of dimension (k,k).

alpha

a numeric vector of length k, which regulates the shape of the density.

fixed.comp

a vector containing a subset of 1:k which selects the components whose values are to be fixed; it must be of length k-2.

fixed.values

a numeric vector of values taken on by the components fixed.comp; it must be of the same length of fixed.comp.

n

an integer value which determines the grid size of the density computations and plot.

Value

• A list containing the following elements:
• cumulants
  two lists as returned by msn.conditional.
• pdf
  a list containing the coordinates x and y of the points where the densities have been evaluated, and the matrices f.exact and f.fitted of the exact and fitted conditional densities.
• rel.error
  summary statistics of relative and absolute error of the approximation.

Side Effects

A contour plot of the exact and approximate densities is produced on a graphical device.

References

Azzalini, A. and Capitanio, A. (1999). Statistical applications of the multivariate skew-normal distribution. J.Roy.Statist.Soc. B 61, 579--602.

See Also

msn.conditional, dmsn

Examples

Omega <- diag(3)+0.5*outer(rep(1,3),rep(1,3))
a<- msn.cond.plot(rep(0,3), Omega, 1:3, 3, -0.75)