bothsidesmodel.mle {msos} | R Documentation |
Calculate the maximum likelihood estimates
Description
This function fits the model using maximum likelihood. It takes an optional
pattern matrix as in (6.51), which specifies which
's
are zero.
Usage
bothsidesmodel.mle(x, y, z = diag(qq), pattern = matrix(1, nrow = p, ncol = l))
Arguments
x |
An |
y |
The |
z |
A |
pattern |
An optional |
Value
A list with the following components:
- Beta
The least-squares estimate of
.
- SE
The
matrix with the
th element being the standard error of
.
- T
The
matrix with the
th element being the
-statistic based on
.
- Covbeta
The estimated covariance matrix of the
's.
- df
A
-dimensional vector of the degrees of freedom for the
-statistics, where the
th component contains the degrees of freedom for the
th column of
.
- Sigmaz
The
matrix
.
- Cx
The
residual sum of squares and crossproducts matrix.
- ResidSS
The dimension of the model, counting the nonzero
's and components of
.
- Deviance
Mallow's
Statistic.
- Dim
The dimension of the model, counting the nonzero
's and components of
- AICc
The corrected AIC criterion from (9.87) and (aic19)
- BIC
The BIC criterion from (9.56).
See Also
bothsidesmodel.chisquare
, bothsidesmodel.df
,
bothsidesmodel.hotelling
, bothsidesmodel.lrt
,
and bothsidesmodel
.
Examples
data(mouths)
x <- cbind(1, mouths[, 5])
y <- mouths[, 1:4]
z <- cbind(1, c(-3, -1, 1, 3), c(-1, 1, 1, -1), c(-1, 3, -3, 1))
bothsidesmodel.mle(x, y, z, cbind(c(1, 1), 1, 0, 0))