Prediction intervals for nlme and gnls models

Share:

Description

Prediction intervals for nlme and gnls models

Details

The main purpose of the package is to provide standard errors for predictions of simple models fitted by nlme or gnls.

The main function is varPredictNlmeGnls using function attachVarPrep.

Further functionality of package twNlme includes
  • Variance of sum of predictions varSumPredictNlmeGnls

  • treating gnls models as special random effects models with zero random effects, i.e accessing fixed (coefficients) and random (none) of gls models: fixef.gls, ranef.gls

  • extracting Variance Covariance matrix of fixed and random effects: varRanef.lme, varFixef.lme for both lme and gls models

  • extracting Standard error of fixed effects: varRanef, seFixef.lme for both lme and gls models

Data
  • tree weights in relation to diameter and height: Wutzler08BeechStem

  • nlme model fit to the tree weight data: modExampleStem

Author(s)

Thomas Wutzler <twutz@bgc-jena.mpg.de>