spaMM (version 1.7.2)

welding: Welding data set

Description

The data give the results of an unreplicated experiment for factors affecting welding quality conducted by the National Railway Corporation of Japan (Taguchi and Wu, 1980, cited in Smyth et al., 2001). It is a toy example for heterocedastic models and is also suitable for illustrating fit of overparameterized models.

Usage

data("welding")

source

The data were downloaded from http://www.statsci.org/data/general/welding.txt on 2014/08/19 and are consistent with those shown in table 5 of Bergman and Hynén (1997).

References

Bergman B, Hynén A (1997) Dispersion effects from unreplicated designs in the $2^{k-p}$ series. Technometrics, 39, 191–98. Smyth GK, Huele AF, Verbyla AP (2001). Exact and approximate REML for heteroscedastic regression. Statistical Modelling 1, 161-175. Taguchi G, Wu Y (1980) Introduction to off-line quality control. Nagoya, Japan: Central Japan Quality Control Association.

Examples

data(welding)
## toy example from Smyth et al.
HLfit(Strength ~ Drying + Material,resid.formula = ~ Material+Preheating ,data=welding)
## toy example of overparameterized model
HLfit(Strength ~ Rods+Thickness*Angle+(1|Rods),resid.formula = ~ Rods+Thickness*Angle ,data=welding)