HW 2 Key
Part a
For the first experiment
= + +
25
1
4 20
For the second experiment
= + +
18
1
4 13
Part b
data ex1;
length trt$ 15;
input trt$ hen y;
cards;
Premolt
1
92.57
.
;
proc sort data=ex1;
by trt;
run;
proc print data=ex1;
run;
proc mixed data
Chapter 12 Fractional Factorial Designs
Lets consider a complete factorial of a 27. Here is the model for a
single replicate. Note that I have not used subscripts just to save time.
Y = + A + B + AB + C + AC + BC + ABC + D + AD + BD + ABD + CD + ACD + BCD
Chapter 11 Incomplete Block Designs: Factorial Treatment Designs
The 2n Factorial Design
2n has n factors and each is at 2 levels e.g. A0 A1, B0 B1, C0 C1 is a 23
3n has n factors and each is at 3 levels e.g. A0 A1 A2, B0 B1 B3 is a 32
These are often use
Chapter 15 - Analysis of Correlated Measures
Repeated Measures in either Time or Space
Change over time or space is an important aspect of many experiment.
It is far more efficient and less costly to make repeated observations
on the same experimental uni
Chapter 16 Crossover Designs
In a crossover study the treatments are administered in sequence to
each experimental unit. A treatment is administered for a specific
period of time after which another treatment is administered to the
same unit. The treatmen
Chapter 1 Research Design Principles
The Legacy of Ronald A. Fisher
http:/en.wikipedia.org/wiki/Ronald_Fisher
I occasionally meet geneticists who ask me whether it is true that the great
geneticist R.A. Fisher was also an important statistician.
L.J. Sava
Chapter 3 - Treatment Comparisons
There are implicit questions in most experiments. Sometimes those
are easy to answer sometimes they are not. Lets assume that we
wanted to talk about the means of the treatments from the meat
problem. If we just want to k
Chapter 5 - Random Effects Models for Variances
Traditional definitions
Fixed - treatment level is reproducible, our interest is the means and
our inference is to just the levels tested.
Random - treatment level is a random draw from a larger population,
Chapter 4 Diagnosing Agreement Between the Data and the Model
Valid Analysis Depends on Valid Assumptions
NID 0, 2
For valid estimates and tests of hypotheses we have to meet the above
critical assumptions regarding the errors which we explain as:
1) The
Chapter 7. Factorial Treatment Designs: Random and Mixed Models
A manufacturer was developing a new spectrophotometer for use in
medical clinical laboratories. The development process was at the
pilot stage of assembly after which machine performance was
Chapter 6 - Factorial Treatment Designs
Factorial Treatment Designs allow us to look at how multiple
treatment factors affect our response variable AND how the
treatments interact to affect our response variable. This is all about
interactions. The follow
Chapter 8 - Complete Block Designs (RCBD)
In the designs we have looked at so far we have assumed that we have
sufficient homogeneous experimental units to complete an
experiment. What if we dont have that many? One approach is to use
some type of blockin
Chapter 9 - Incomplete Blocks
These are used when you dont have sufficient experimental units to
run a complete replication in a block. Or we may have so many
treatments that the block does not reduce our experimental error
variance and thus is not effect
Chapter 10 - Incomplete Block Designs: Resolvable and Cyclic Designs
Resolvable designs - Blocks are grouped so that each group of blocks
constitutes one complete replication of the treatments. These were
first used by plant breeders, but have since been
Chapter 17 Analysis of Covariance
Analysis of covariance can be summarized as a technique to reduce
experimental error by utilizing additional factor(s) which are not part
of the treatments but were in place prior to imposing treatments and
are affecting
Mean and Variance of a Linear Function
1) Introduction: In previous courses you have been taught many
definitions for means and variances. For example,
Y
2
Y
2
n
Y
2
Y Y 2
2
2
Y Y 2
1
2
n
n
2
2
Yi Yi 2 2 Yi Yi 2 2
1
2
n
The goal of this lecture is t
542 Exam 1
Wednesday 27 February 2013
Name_
Lab_
Question
Points Points
Possible Earned
1
10
2
15
3
15
4
15
5
15
6
15
7
15
Total
100
1
Points missed on this page_
1) (10 points) In the blank column (Col #3) write the equation number (Eq #) that best agree
Practice and Problems for 542 Exam I 2015
All quiz and homework questions are also potential questions on exams
For the following table, place in the the blank column (col #3) the equation number (Eq #) that best
agrees with the description or term in col
542 Exam 1
Wednesday 5 March 2014
Name_
Question
Lab Time_
Points Points
Possible Earned
1
10
2
10
3
22
4
15
5
8
6
20
7
15
Total
100
1
Points missed on this page_
1) (10 points) In the blank column (Col #3) write the equation number (Eq #) that best agree
HW 4 Key
Problem 1
The first thing that we have to do is that we have to decide how we are going to group the
variances together. In general, we will want to group the more similar variances together.
Another good rule of thumb to remember from 440 is tha
542 Quiz 1 KEY
23 January 2015
k
L ciYi c1Y1 c2Y2 . ckYk
i 1
k
L E( L) ci i c1 1 c2 2 . ck k
i 1
k
E ( L L ) c 2
2
L
2
i 1
2
i
2
i
k
k
c c
i 1,i i i
i i
cov(Yi , Yi )
Assume an experiment has six treatments, it is in a CRD and has nine replications. Yo
542 Quiz 2 KEY
30 January 2015
a) (5 points) Assume an experiment has four treatments, it is in a CRD and has four replications.
1
You want know the average of the first three means i.e. (Y1 Y2 Y3 ) . Show the Linear function, the
3
mean of that linear fu
542 Quiz 4 KEY
13 February 2015
Write the linear model with all degrees of freedom for an experiment with a fixed factor, , with four
levels and five replications in a CRD and thus 20 experimental units. From each experimental unit two
blood samples were
542 Quiz 3 KEY
6 February 2015
CI Estimate t critical , ,MSEdf Estimate
2
These are the same data that you had in your homework.
Experiment 2 This has missing data
Hen Premolt Fasting 60g Bran 80g Bran
1
92.57
Died
127.22
103.20
2
99.95
117.06 Died
100.1