Published 2012 | Version v1
Publication

From Markov moves in contingency tables to linear model estimability.

Description

The aim of this work is to highlight some interesting connections between contingency tables analysis and design of experiments. In particular, we consider two-way tables in correspondence to two-factor designs. We provide a condition that characterizes the estimability of the independence model for all saturated fractions.

Additional details

Identifiers

URL
http://hdl.handle.net/11567/717973
URN
urn:oai:iris.unige.it:11567/717973