next up previous index
Next: Making the ``Data Modeling: Up: Linear Models Previous: Linear Models with Multiple   Index

Click for printer friendely version of this HowTo


In conclusion, when viewing linear models from a matrix perspective, you build the model to mimic the physics or physiology that you believe to be operational - including terms that dominate the processes - and you're not stuck with a recipe book of 1-way ANOVA, ANCOVA etc. The matrix approach to linear modeling simply suppresses the mechanics of the computational process and lets you focus on the science - the model of what you think is going on (the process).

Frank Starmer 2004-05-19