h: Multivariate t distribution with missing data
In emilygoren/MixtClust: Robust Clustering for Complete and Incomplete Data

h	R Documentation

Multivariate t distribution with missing data

Description

Compute the marginal density of the multivariate t distribution for the observed coordinates.

Usage

h(x, mu, sigma, nu, grp, Ru)

Arguments

`x`	Numeric. A vector (of length p) or matrix (with p columns). Missing values for row i correspondind to unique pattern of missingness m are indicated by the m^{th} element of the selection index list `D`.
`mu`	Numeric. A vector of length p representing the mean.
`sigma`	Numeric. A p \times p non-negative definite matrix (or its Cholesky decomposition).
`nu`	Numeric. A postitive scalar.
`grp`	Numeric. A vector of length n with elements indicating the which row of `Ru` to use as the observed coordinates for each of the n observations.
`Ru`	Binary matrix. Each row corresponds to a unique pattern of missingness, where 1 indicates observed and 0 indicates missing coordinates.

Value

The multivariate t density for all rows in x using degrees of freedom nu, mean vector mu, covariance matrix sigma, and observed coordinates specified by rows in Ru.