Influence analysis in two-treatment cross-over designs with special reference to the ABBA|BAAB design

Sammanfattning

This work is to develop methodology to detect influential observations in linear mixed model for multiple-period two-treatment cross-over designs. Existence of explicit maximum likelihood estimates (MLEs) of variance parameters as well as of mean parameters in the mixed model with treatment, residual, period and sequence effects is proven. Special reference is taken to the four-period ABBA|BAAB design. Case-weighted perturbations are performed. The influence quantities on each parameter estimate and their dispersion matrix are presented as closed-form functions of residuals in the unperturbed model.

Nyckelord

delta-beta influence; explicit maximum likelihood estimate; mixed linear model; multiple-period cross-over design; perturbation scheme; variance-ratio influence

Publicerad i

Research report - Department of Statistics, University of Stockholm
2011, nummer: 2011:2
Utgivare: Department of Statistics, Stockholm University

SLU författare

• von Rosen, Dietrich

  • Institutionen för energi och teknik, Sveriges lantbruksuniversitet
    
  • Linköpings universitet

UKÄ forskningsämne

Sannolikhetsteori och statistik

Permanent länk till denna sida (URI)

https://res.slu.se/id/publ/37297