condMom {bayesm} R Documentation

## Computes Conditional Mean/Var of One Element of MVN given All Others

### Description

`condMom` compute moments of conditional distribution of the ith element of a multivariate normal given all others.

### Usage

`condMom(x, mu, sigi, i)`

### Arguments

 `x` vector of values to condition on; ith element not used `mu` mean vector with `length(x)` = n `sigi` inverse of covariance matrix; dimension n x n `i` conditional distribution of ith element

### Details

x ~ MVN(mu, sigi^{-1}).

`condMom` computes moments of x_i given x_{-i}.

### Value

A list containing:

 `cmean ` conditional mean `cvar ` conditional variance

### Warning

This routine is a utility routine that does not check the input arguments for proper dimensions and type.

### Author(s)

Peter Rossi, Anderson School, UCLA, perossichi@gmail.com.

### References

For further discussion, see Bayesian Statistics and Marketing by Rossi, Allenby, and McCulloch.
http://www.perossi.org/home/bsm-1

### Examples

```sig  = matrix(c(1, 0.5, 0.5, 0.5, 1, 0.5, 0.5, 0.5, 1), ncol=3)
sigi = chol2inv(chol(sig))
mu   = c(1,2,3)
x    = c(1,1,1)

condMom(x, mu, sigi, 2)
```

[Package bayesm version 3.1-4 Index]