ESSgls: Effective Sample Size for 'gls' Parameters
In PhilipPallmann/SimLongi: Simultaneous Inference in Longitudinal Data Settings

Description Usage Arguments Details Value Note Author(s) References See Also Examples

Compute the effective sample sizes associated with the parameter estimates of a general linear model fitted via gls.

1	ESSgls(model, IDcolumn)

`model`	A general linear model fit of class `gls` from `nlme`.
`IDcolumn`	An integer giving the number of the column in the `data.frame` on which the model fit is based that contains the ID variable (cluster identifier).

The data.frame used for model fitting must be sorted along the ID variable, otherwise you might get mock results!

A vector giving the effective sample sizes associated with the estimated parameters of the gls model.

Models with multiple hierarchical levels are currently not supported.

Philip Pallmann pallmann@biostat.uni-hannover.de

Faes, C., Molenberghs, G., Aerts, M., Verbeke, G., Kenward, M. G. (2009) The effective sample size and an alternative small-sample degrees-of-freedom method. The American Statistician, 63(4), 389–399.

ESSlme

### Using the example from the gls help page
library(nlme)
data(Ovary)
gls1 <- gls(follicles ~ sin(2 * pi * Time) + cos(2 * pi * Time), Ovary,
            correlation=corAR1(form=~1|Mare))
#ESSgls(model=gls1, IDcolumn=1)