clme_resids {CLME} | R Documentation |
Computes several types of residuals for objects of class clme
.
clme_resids(formula, data, gfix = NULL)
formula |
a formula expression. The constrained effect(s) must come before any unconstrained covariates on the right-hand side of the expression. The first |
data |
data frame containing the variables in the model. |
gfix |
optional vector of group levels for residual variances. Data should be sorted by this value. |
For fixed-effects models Y = X*b + e, residuals are given as ehat = Y - X*betahat. For mixed-effects models Y = X*b + U*xi + e, three types of residuals are available. PA = Y - X*betahat\ SS = U*xihat\ FM = Y - X*betahat - U*xihat
List containing the elements PA
, SS
, FM
, cov.theta
, xi
, ssq
, tsq
.
PA
, SS
, FM
are defined above (for fixed-effects models, the residuals are only PA
). Then cov.theta
is the unconstrained covariance matrix of the fixed-effects coefficients, xi
is the vector of random effect estimates, and ssq
and tsq
are unconstrained estimates of the variance components.
There are few error catches in these functions. If only the EM estimates are desired,
users are recommended to run clme
setting nsim=0
.
By default, homogeneous variances are assumed for the residuals and (if included)
random effects. Heterogeneity can be induced using the arguments Nks
and Qs
,
which refer to the vectors (n1, n2 ,... , nk) and
(c1, c2 ,... , cq), respectively. See
CLME-package
for further explanation the model and these values.
See w.stat
and lrt.stat
for more details on using custom
test statistics.
## Not run: data( rat.blood ) cons <- list(order = "simple", decreasing = FALSE, node = 1 ) clme.out <- clme_resids(mcv ~ time + temp + sex + (1|id), data = rat.blood ) ## End(Not run)