Chapter 4 Notes
Brief Review
Before we discuss the ANOVA model, lets a recall the classification-based testing we have done in the
previous chapters.
In Chapter 3 we talked about categorical data analysis. We had observations classified (or categorized)
b
SAS Overview Notes (Chapter 1)
It is probably more instructive to see the interface and language by example, but we will summarize a
few main points for future reference.
The Interface
Editor Window - where you write the code
Log Window where notes and er
Chapter 3
Simple Inference for Categorical
Data
Categorical Data
For a single variable:
Just look at counts
Number in Group A, B, C, etc.
Could use graphics like pie and bar charts
Can infer proportions of the groups in the
broader population
Example:
Chapter 16
Principal Components Analysis
Review: Previous Models
Response and a bunch of predictors
Strong correlations between predictors are a
problem
Hope to use a small number of fairly unrelated
predictor variables
Highly Correlated Data
Common
Chapter 4
Analysis of Variance
(Balanced Case)
Review: Categorical Data
In Chapter 3:
Combinations of categorical variables defined
cells
Had counts for each cell
Looked at association between categorical
variables
Review: Two-Sample T-Test
Classificat
Chapter 16 Notes
Principal Component Analysis
The General Idea and Theory
Its common to have data sources with highly related variables, for instance:
Databases with many variables on similar product characteristics
Survey results where questions are high
Homework 1
Due: Tuesday June 18 at 7pm
See general homework tips and submit your files via the course website.
For all exercises, use the iris data set from the SAS help (e.g. data=sashelp.iris).
Exercise 1:
a) Obtain box plots for petallength by species