Biometrical Letters vol. 46(2), 2009, pp. 113-128
|
ESTIMATION AND HYPOTHESIS TESTING IN BLOCK DESIGNS WITH NESTED ROWS AND COLUMNS Agnieszka Łacka1, Maria Kozłowska1, Barbara Bogacka2 1Department of Mathematical and Statistical Methods, Poznań University of Life Sciences, Wojska Polskiego 28, 60-637 Poznań, Poland, e-mail: aga@riders.pl, e-mail: markoz@up.poznan.pl 2Queen Mary, University of London, School of Mathematical Sciences, Mile End Road, London E1 4NS, UK, e-mail: b.bogacka@qmul.ac.uk |
We present methods of estimation and hypothesis testing for treatment contrasts in generally balanced block designs with nested rows and columns. The linear mixed effects model is built by randomization of all the nuisance effects, that is blocks, rows and columns. This allows for model decomposition into five sub-models according to decomposition of the random sample space into five strata. We present methods for estimation of the treatment contrasts, for testing within the strata and also various methods of combining the strata information. In this work we largely apply the results given by Caliński and Kageyama (2000, 2003).
block design with nested rows and columns, combined estimator, combined test, general balance
