Chapter 1: Linear Regression with One Predictor Variable
1.3 Simple Linear Regression Model with Distribution of Error Term Unspecified
Model: Yi 0 1 X i i
(i = 1, 2, , n)
(1.1)
Yi : the value of the response variable (DV) for the i-th subject
0 and 1 :
Chapter 9: Building the Regression Model I: Model Selection and Validation
9.2 Surgical Unit Example
A hospital surgical unit was interested in predicting survival in patients undergoing a particular
type of liver operation. A random selection of 54 patie
Chapter 8 Regression Models for Quantitative and Qualitative Predictors
8.1 Polynomial Regression Models
One Predictor Variable Second Order
Yi 0 1 xi 11 xi2 i
(8.2)
where: xi X i X
One Predictor Variable Third Order
Yi 0 1 xi 11 xi2 111 xi3 i
(8.5)
Note
MATRIX ALGEBRA
PSY613/EDEP605
MATRIX
A matrix is a rectangular array of elements:
1 2 3 4
4 5 6 9
DIMENSIONS (OF THE MATRIX)
4X3
(# OF ROWS) X (# OF COLUMNS)
VARIABLES
10
12
SUBJECTS
13
16
4
6
2
8
18
21
20
16
xij
SCORE OF ROW (SUBJECT) i ON COL
Chapter 6 Multiple Regression I
6.1 Multiple Regression Model
Need for Several Predictor Variables: An experimenter typically will wish to
investigate a number of predictor variables simultaneously because almost always more
than one key predictor variabl
Chapter 16: Single-Factor Studies
16.1 Single-Factor Experimental and Observational Studies
Read Examples 1-3 (pp. 677-678)
16.2 Relation between Regression and Analysis of Variance
(1) The explanatory or predictor variables in ANOVA models may be qualita
Chapter 7 Multiple Regression II
7.1 Extra Sums of Squares
Definition: Extra Sum of Squares (2 Predictors): Suppose that there are two predictors
X1 and X2.
(i) Suppose that X1 is the extra variable. Then the extra sum of square is:
SSR(X1|X2) = SSR(X1,X2
Chapter 17: Analysis of Factor Level Means
17.1 Introduction
In Chapter 16, we discussed the F test for determining whether or not the factor level
means i differ. When this test leads to the conclusion that the factor level means i are
equal, no further
Chapter 23: Two-Factor Studies with Unequal Sample Sizes
Two-factor studies frequently involve unequal treatment or cell sample sizes for a
variety of reasons. In observational studies, the investigator often has little or no control
over the cell sample
Chapter 27: Repeated Measures and Related Designs
27.1 Elements of Repeated Measures Design
1. Description of Designs: Repeated measures designs utilize the same subject (person,
store, plant, test market, etc.) for each of the treatments under study. The
Chapter 19: Two-Factor Studies (Two-Way ANOVA) with Equal Sample Sizes
19.1 Two-Factor Observational and Experimental Studies
1. See Examples 1, 2, and 3 (pp. 812-814)
Example 1: A company investigated the effects of selling price (55, 60, 65) and type of
Shereen Cohen
PSY611/EDEP603 Chapter 8 Questions
A. Polynomial and interaction regression models
1. What do you mean by a polynomial regression model? Has a polynomial term,
meaning a term with an exponent, and is therefore not linear.
2. Why is it import
Repeated-Measures Analysis
The repeated-measures designs are the designs in which subjects are
measured on the same variable several times.
Example 1: Matched-pairs (dependent-sample) t-test
Subjects
1
2
s
Pretest scores
Y11
Y12
Y1s
Treatment
Posttest sco
Shereen Cohen
PSY611/EDEP603 Chapter 9 Questions
1. What is the relationship between choosing a model with the highest value of R
square and choosing a model with the lowest value of SSE (error sum of
squares)? R2 = SSR/SST = 1 (SSE/SST), so when SSE is s
Shereen Cohen
Homework 1
PSY611
A. Simple regression and expected value
1.
What do you mean by the simple regression? One predictor
2.
What components constitute the equation for a simple regression? Dependant variable,
predictor, error, constant (Y inter
PSY611/EDEP603
Design and Analysis of Experiments
Spring 2014
Instructor: Kentaro Hayashi, PhD
Office: Sakamaki Hall D-403
Phone: (808) 956-2846 (Please send me an email message, rather than leaving a voice
mail.)
Emails: hayashik@hawaii.edu (Email is the
SequentialSS(TypeISS)vs.PartialSS(TypeIIISS)
BasedonLittle,R.C.,Freund,R.J.,&Spector,P.C.(1991).SASsystemforlinearmodels (3rded.).Cary,
NC:SASInstitutes.,withsomerevisionsbyKH
The Type I SS are commonly called sequential sums of squares. They represent a
Shereen Cohen
1/28/14
PSY611/EDEP603 Chapter 2 Questions
A. Power and non-centrality parameter
1. What is the definition for power? The likelihood that you will correctly reject the null
hypothesis.
2. Power is defined given a non centrality parameter. De
Shereen Cohen
PSY611/EDEP603 Chapter 15 Questions on the Design of Experiment
1. What (five) components constitute the design of experiment?
(i) The set of explanatory factors included in the study
(ii) The set of treatments included in the study
(iii) Th
Shereen Cohen
3/18/2014
PSY611/EDEP603 Chapter 17 Questions
1. What are (simultaneous) multiple comparison procedures (also called posthoc tests)?
What we do after the ANOVA with the control of overall alpha level, in order to
determine WHERE the differen
Shereen Cohen
4/8/2014
PSY611/EDEP603 Chapter 19 Questions
Note: Unless specified otherwise, assume two factor studies (i.e., twoway ANOVA).
1. What are the three advantages of a (crossed) multifactorial ANOVA over oneway
ANOVA?
It allows you to compare m
Shereen Cohen
4/15/2014
PSY611/EDEP603 Chapter 23 Questions
Note: Assume that cell sample sizes are unequal in the following questions.
1. What is the name of the approach for testing for twoway ANOVA with unequal sample
sizes, according to our textbook (
Shereen Cohen
PSY 611/ EDEP603 Chapter 6 Questions
1. What do you mean by the multiple regression? In multiple regression, you have
multiple predictors.
2. What is the meaning of regression coefficient 1 associated with X1? The
parameter 1 indicates the c
Shereen Cohen
3/11/14
PSY611/EDEP603 Chapter 16 Questions
Note: All the questions below assume oneway ANOVA to be appropriate.
1. How does the cell means model partition the dependent variable Y?
yij = i + ij
(Mean plus error)
2. Suppose there are two obs
Chapter 22: Analysis of Covariance (ANCOVA)
Analysis of covariance (ANCOVA) is a technique that combines features of analysis of
variance and regression. It can be used for either observational studies or designed
experiments. The basic idea is to augment