STAT 371 Assignment #4
Due Tues. Nov. 29th at noon (12:00 pm) in drop box 14 outside MC 4066 in appropriate STAT 371 slot
SOLUTIONS ( /70)
1) In the midterm you were introduced to the company dataset containing financial data of 266 U.S. Companies
on Sale
Supplementary Review Questions STAT 371 Midterm
1) Janice, the product manager of the nutritious breakfast cereal ChocoSugar Flakes, has called you into her
office. She needs your help in interpreting the output from a regression analysis on the effect o
Some additional review questions to assist in your understanding of concepts covered in STAT 371
1) Consider the audit output we have been discussing in lectures throughout the term, and the associated plot
of the hat values:
Residuals:
Min
1Q
Median
3Q
M
STAT 371 Assignment #4
Due Tues. Nov. 29th at noon (12:00 pm) in drop box 14 outside MC 4066 in appropriate STAT 371 slot
1) In the midterm you were introduced to the company dataset containing financial data of 266 U.S. Companies
on Sales ($ Millions), C
STAT 371  Assignment #1
Yuheng Yu
ID: 20551359
Comment:
There is no significant relationship between size and overhead;
There is no significant relationship between age and overhead;
There is no significant relationship between employees and overhead;
STAT 371 (F16) Midterm Information
Time:
Thursday, Oct. 20. 2:30 3:45 in MC 2065.
Format:
Closed book. Short answer. 75 minutes.
Aides:
Calculator.
Content: Midterm content will be based on material covered in lectures up to and including additional sum o
Stat 371 Final Exam Information F16
Time and Location: See Official Waterloo Examination Schedule
Format: Closed Book. Short answer. Same format as midterms.
Study Aides Allowed: Nonprogrammable calculator. Probability tables will be provided.
Content: C
University of Waterloo
Department of Statistics & Actuarial Science
STATS 331/ Spring 2017
Practice Questions (2)
1. The following matrices were computed for a certain regression problem:
15
= 3626
XX
44428
3626
1067614
11419181
44428
1.2463484
1 =
114
University of Waterloo
Department of Statistics & Actuarial Science
STATS 331/ Spring 2017
Practice Questions (3)
1. Consider the multiple regression model:
y = X +
where the usual assumption have been made, and where is a vector of 5 coefficients.
(a) S
Stat 938: Statistical Consulting
Review
June 29, 2017
Contrasts
A contrast is a linear combination of the parameters of the model
of the form
a
ci i
a
where
i 1
n c
i i
i 1
0
The estimate, C, the SSC and variance of the contrasts are
2
a
C ci yi .
i 1
Stat 430/830
Lecture June 15, 2017
Outline
Fold over Resolution III review
Fold ovre Resolution IV
PlackettBurman Designs
1
Resolution III Designs
(Main Effects aliased with second order effects)
Saturated Designs
When k= N1 then the design is called
Tutorial July 10, 2017
1. A manufacturing engineer is studying the dimensional variability of a particular component that
is produced on three machines. Each machine has two spindles, and four components are
randomly selected from each spindle. The result
Tutorial July 14, 2017
1.
A process engineer is testing the yield of a product manufactured on three machines.
Each machine can be operated at two power settings. Furthermore, a machine has three
stations on which the product is formed. An experiment is c
Chapter 6 Solution
6.12. An article in the AT&T Technical Journal (March/April 1986, Vol. 65, pp. 3950) describes the application of
twolevel factorial designs to integrated circuit manufacturing. A basic processing step is to grow an epitaxial layer
on
Tutorial June 26, 2017
41
1. An unreplicated 2 fractional factorial experiment with four center points was run. The
experimenter used the following factors:
Factor
Natural levels
A time
B temperature
C concentration
D pressure
10, 50 (minutes)
200, 300 (
STAT 431, Spring 2017
STAT 431: Generalized Linear Models and Their Application
Tuesday & Thursday, 2:303:50, DC 1350
Professor
Leilei Zeng
Office Hours: Tue. & Thu. 4:005:30, M34223
or by appointment, [email protected]
Teaching Assistants
T
Log Linear Models for 3Way Contingency
Tables
91 / 113
Log Linear Models for 3Way Contingency Tables
Associations of V , W and Z and Corresponding Loglinear Models
Type of Association

Null Hypothesis

Loglinear Models
(VWZ )
Homogeneous Association
Stat 430/830
Random Effects
Lecture June 27, 2017
Outline
Expected MSE
Multifactorial designs with random factors
Approximate Ftests
1
Two Random Factors
Finding Expected Mean Squares
y i j ij k (ij )
Effect
i
j
k
E(MS)
i
j
ij
(ij)k
Alive: subscript
Stat 430/830: Analysis of Variance
Lecture May 2, 2017
Outline
Main concepts in design of experiments
Single factorial experiment
Model and estimation
Analysis of variance
1
Introduction
Methods for comparing two populations, or
two formulations or two l
Solutions Chp 13 November 4, 2016
13.3 Reconsider the data in Problem 5.8. Suppose that both factors, machines and operators, are
chosen at random.
(a) Analyze the data from this experiment.
Machine
Operator
1
1
109
110
2
110
115
3
108
109
4
110
108
2
110
Stat 430/830
Lecture June 8, 2017
Outline
Blocking in factorial designs
Confounding
Partial confounding
Fractional Factorials
Definition
How to construct them
Aliasing
1
Blocking
Used to remove effects of nuisance
factors
Induce randomization res
Chapter 4 Solutions
4.1.
The ANOVA from a randomized complete block experiment output is shown below.
Source
DF
SS
MS
F
P
Treatment
4
1010.56
?
29.84
?
Block
?
?
64.765
?
?
Error
20
169.33
?
Total
29
1503.71
(a) Fill in the blanks. You may give bounds on
Stat 430/830
Lecture June 15, 2017
Outline
More examples of Fractional Factorials
Blocking in Fractional Factorials
Regular and Nonregular designs
Fold over
Blocking in 2kp Designs
PlackettBurman Designs
1
Example 262
Injection molding process. Respons
Stat 430/830
Lecture June 6, 2017
Outline
Prediction SS
A single replicated or unreplicated 2k
design
Optimal Design
Addition of center point & augmented
designs
1
Prediction Sum of Squares
In matrix notation the model can be expressed as
Ynx1 X nxp
Stat 431  ASSIGNMENT 4
Spring 2017
Due: July 24 (Monday), 2017
Reminder:
 Your assignment must be submitted by 11:00 pm on the due date via Crowdmark. You will
receive an email with a personal submission link (if not, check your spam folder!).
 Be sure
Stat 430/830
Lecture May 11, 2017
Outline
Randomized block designs
Nuisance factor
Controllable and uncontrollable factors
Blocking
Randomization restriction
Interaction
Latin square designs
1
Example
Consider an experiment to
compare four hormone
Stat 430/830: Analysis of Variance
Lecture May 4, 2017
Outline
Model and estimation
ANOVA as Regression
Model Checking
Treatment comparison
1
Analysis of Variance
The main idea of ANOVA is to compare the
between and the within variability
If between > w
Stat 430/830
Randomized Block Designs
Factorial Designs
Lecture May 25, 2017
Outline
Goodness of fit test
Blocking in factorial designs
Three factors factorial design
Sample size estimation
1
Lack of Fit and Pure Error Sum of
Squares
Consider an experime
Stat 430/830
Factorial Designs
Lecture May 18, 2017
Outline
Review of the twofactor factorial
model for fixed effects and its ANOVA
table
Introduction to response surface
The regression model for the two factor
factorial
Estimable functions
Factoria
Stat 430/830
Lecture July 11, 2017
Outline
Split plot designs review
In adverted split plot designs
Splitsplit plot designs
Analysis of Covariance
1
SplitPlot Design
May consider the experiment was run in two stages:

Stage 1 or whole plot: randomize m