Fitted gnls Object
An object returned by the gnls function, inheriting from class "gnls" and also from class "gls", and representing a generalized nonlinear least squares fitted model. Objects of this class have methods for the generic functions anova, coef, fitted, formula, getGroups, getResponse, intervals, logLik, plot, predict, print, residuals, summary, and update.
The following components must be included in a legitimate "gnls"
object. - apVar: an approximate covariance matrix for the variance-covariance coefficients. If apVar = FALSE in the control values used in the call to gnls, this component is equal to NULL.
call: a list containing an image of the gnls call that produced the object.
coefficients: a vector with the estimated nonlinear model coefficients.
contrasts: a list of the contrast matrices used to represent factors in the model formula. This information is important for making predictions from a new data frame in which not all levels of the original factors are observed. If no factors are used in the model, this component will be an empty list.
dims: a list with basic dimensions used in the model fit, including the components N - the number of observations used in the fit and p - the number of coefficients in the nonlinear model.
fitted: a vector with the fitted values.
modelStruct: an object inheriting from class gnlsStruct, representing a list of model components, such as corStruct and varFunc objects.
groups: a vector with the correlation structure grouping factor, if any is present.
logLik: the log-likelihood at convergence.
numIter: the number of iterations used in the iterative algorithm.
plist:
pmap:
residuals: a vector with the residuals.
sigma: the estimated residual standard error.
varBeta: an approximate covariance matrix of the coefficients estimates.
José Pinheiro and Douglas Bates bates@stat.wisc.edu
gnls, gnlsStruct