As can be seen from the iterative proportional fitting algorithm, the ms for the model of no

As can be seen from the iterative proportional fitting algorithm, the ˆm’s for the model of no three-factor interaction depend only on the 3 two-dimensional marginal tables. Discuss how this fact can be used to develop a more complete analysis for the Gilby data of Exercise 2.7.3.

What assumptions must be made and what techniques should be used?

What problems will Standard VIII cause?

Project 3.8.13. Write a computer program to fit the model of no threefactor interaction to the Gilby data of Exercise 2.6.3.

This can be done using iterative proportional fitting. Assuming that this model fits, do any submodels fit adequately?