Outline
Topic 26 - Two-Way
Two-way Random eects
Random Eects
Data
Model
STAT 525
Variance component estimation
Professor Bruce Craig
F-tests
26
26-1
Example Page 1080
Data for Two-Way Design
Int
Outline
Poisson Regression
Topic 30 - Poisson
Regression and Categorical
Background
Model
Data Analysis
Inference
Categorical Data Analysis
STAT 525
Goodness of t test
Professor Bruce Craig
Tes
STAT525 HOMEWORK#3
Due February 4, 2016, BEFORE class
1. KNNL Problem 2.57.
2. KNNL Problem 2.61.
3. KNNL Problem 3.19
4. KNNL Problem 3.20
5. (SAS Exercise) Use the crime rate data described in KNNL
STAT525 HOMEWORK#2
Due January 28, 2016 BEFORE class
1. KNNL Problem 2.53
2. In four sentences or less, explain what a sampling distribution is and why it is important
to know the sampling distributio
EXAM III
STAT 525
Name
STAT 525
(5 pts)
Problem Points Score
1
20
2
25
3
25
4
25
USE YOUR TIME WISELY
GIVE CLEAR, CONCISE ANSWERS
SHOW YOUR WORK TO RECEIVE PARTIAL CREDIT
WRITE LEGIBLY. ANYTHING UNREA
STAT525 HOMEWORK#11 Solution
1. KNNL Problem 24.12
NA.
2. KNNL Problem 24.13
NA.
3. KNNL Problem 24.14
NA.
4. KNNL Problem 25.3
In (1), these are the only three 8 hour shifts so it would be a xed eect
STAT525 HOMEWORK#1
Due January 21, 2016 BEFORE class
1. KNNL 1.3
2. KNNL 1.5
3. KNNL 1.22
4. KNNL 1.26
5. A regression analysis of a set of data produced the following fitted equation: yb = 3+8x.
(a)
STAT525 HOMEWORK#8 Solutions
1. KNNL Problem 17.2
It is true that the standard errors and estimates do not change when using the Bonferroni
adjustment. What does change, however, is the t statistic an
STAT525 HOMEWORK#9 Solution
1. KNNL Problem 19.5
The Factor B main eects involve the means at each level of B (averaging out A). These
turn out to all be 269. This means the Factor B eects (the s) are
STAT525 HOMEWORK#12 Solution
1. KNNL Problem 25.24
NA.
2. KNNL Problem 25.17 (You can omit part e)
Mixed ANOVA model 25.42 is the restricted mixed model so we need to be careful using
SAS. As far as t
STAT525 HOMEWORK#10 Solution
1. KNNL Problem 23.1
(a) No. By randomly discarding cases, she is, on average, going to end up with samples
with the same mean as the original data. Consequently, there is
STAT525 HOMEWORK#4
Due February 11, 2016, BEFORE class
1. Consider the following data set that describes the relationship between velocity of
an enzymatic reaction (V ) and the substrate concentration
STAT525 HOMEWORK#5
Due February 18, 2016, BEFORE class
1. Use the patient satisfaction data described in KNNL Problem 6.15.
a Compute the pairwise correlations between the Xs and between each X and Y
STAT525 HOMEWORK#6
Due February 25, 2016, BEFORE class
Please note that some homework problems require a lot of SAS analysis and output. You
are partially graded on how your answers are organized, con
STAT525 HOMEWORK#10
Due April 14, 2016, BEFORE class
1. KNNL Problem 23.1
2. KNNL Problem 23.2
3. KNNL Problem 23.7
4. KNNL Problem 23.12
5. KNNL Problem 22.3
6. KNNL Problem 22.7
7. KNNL Problem 22.8
STAT525 HOMEWORK #7
Due March 3, 2016, BEFORE class
1. KNNL Problem 11.7 Hint on the Breusch-Pagan test: regress the log-squared residuals log(e2i ) against Xi by considering log(e2i ) as an estimate
STAT525 HOMEWORK#2 Solution
1. NKNL Problem 2.53
(a) The likelihood function for the linear model is
[
]
n
1
1
L(0 , 1 , ) =
exp
(Yi 0 1 Xi )2 .
(2 2 )n/2
2 2 i=1
2
(b) Taking logarithm on both sides
STAT525 HOMEWORK#8
Due March 24, 2016, BEFORE class
1. KNNL Problem 17.2
2. KNNL Problem 17.3
3. KNNL Problem 17.12 For d, please calculate the confidence intervals used for the plot but ignore the pl
STAT525 HOMEWORK #9
Due April 7, 2016, BEFORE class
1. KNNL Problem 19.5
2. KNNL Problem 19.14
3. KNNL Problem 19.15
4. KNNL Problem 19.28
5. A clay tile company was interested in studying the eects o
STAT525 HOMEWORK#4 Solution
1. Consider the given data set that describes the relationship between velocity of an
enzymatic reaction (V ) and the substrate concentration (C). You are asked to investig
STAT525 HOMEWORK#1 SOLUTION
1. KNNL Problem 1.3
If, for a given elapsed time since the molding process (X), there is no chance variation in
the hardness (Y ), then the relationship is mathematical. In
STAT 525
FALL 2017
Chapter 1
Linear Regression with One Predictor
Professor Dabao Zhang
Goals of Regression Analysis
Serve three purposes
Describes an association between X and Y
In some applicatio
STAT525 HOMEWORK#1
Due August 31, 2017
1. KNNL 1.3
2. KNNL 1.5
3. KNNL 1.22
4. KNNL 1.26
5. A regression analysis of a set of data produced the following fitted equation: yb = 3+8x.
(a) If x increases
STAT 525
FALL 2017
Chapter 3
Diagnostics and Remedial Measures
Professor Dabao Zhang
Diagnostics
Procedures to determine appropriateness of the model and
check assumptions used in the standard infere
STAT 525
FALL 2017
Chapter 7
General Linear Test and Multicollinearity
Professor Dabao Zhang
General Linear Test
Comparison of a full model and reduced model that involves
a subset of full model pred
STAT 525
FALL 2017
Chapter 6
Multiple Regression
Professor Dabao Zhang
The Data and Model
Still have single response variable Y
Now have multiple explanatory variables
Examples:
Blood Pressure vs
STAT 525
FALL 2017
Chapter 5
Matrix Approach to Simple Linear
Regression
Professor Dabao Zhang
Matrix
Collection of elements arranged in rows and columns
Elements will be numbers or symbols
For exa
STAT525 HOMEWORK#6
Due October 12, 2017
Please note that some homework problems require a lot of SAS analysis and output. You
are partially graded on how your answers are organized, concise / to the p
STAT525 HOMEWORK#5
Due September 28, 2017
1. Use the patient satisfaction data described in KNNL Problem 6.15.
a. Compute the pairwise correlations between the Xs and between each X and Y .
Which X va
STAT525 HOMEWORK#2
Due on September 7, 2017
1. KNNL Problem 2.53
2. In four sentences or less, explain what a sampling distribution is and why it is important
to know the sampling distribution of a te
STAT 525
FALL 2017
Chapter 4
Miscellaneous Topics
Professor Dabao Zhang
Simultaneous Inference
Consider a collection of CIs / hypothesis tests
Each interval has (1 )% confidence level
What about th
STAT525 HOMEWORK#3
Due on September 14, 2017
1. KNNL Problem 2.57.
2. KNNL Problem 2.61.
3. KNNL Problem 3.19
4. KNNL Problem 3.20
5. (SAS Exercise) Use the crime rate data described in KNNL Problem 1
STAT525 HOMEWORK#4
Due on September 21, 2017
1. Consider the following data set that describes the relationship between velocity of
an enzymatic reaction (V ) and the substrate concentration (C).
Conc