8/11/2019 1112SEM1-ST5203
1/14
NATIONAL UNIVERSITY OF SINGAPORE
EXAlVIIN
ATION
ST52 3
EXPERIMENT L DESIGN
(Semester 1 : AY 2011-2012)
Novernber
2011
- Time Allowed: 2 Hours
INSTRUCTIONS
TO C NDID TES
1.
This examination paper contains
FIVE (5)
questions and comprises
FOURTEEN
(14)
printed pages (including the present page).
2.
Candidates must answer ALL questions on this paper. The total mark for this paper
is
60.
3. Non-programmable calculators can be used.
4.
This is an OPEN BOOK examination.
5. Write down your matriculation number and seat number in the space below.
Matriculation No: _
Seat No: _
Question Points scored
Max. points
1
12
2
12
3
12
12
5
12
Total
60
1
8/11/2019 1112SEM1-ST5203
2/14
ST5203
1. (12pts)Anagronomistwants
to
conductafield trialtocomparethe yieldofthreevarieties
oftomatoin
the
following9plots.
1
2
3
4
5
7 8
91
(a) (4pts)Supposethis
is
theonlyinformationwe have,whichdesigndoyousuggest? ow
shouldtheagronomistdotherandomization?
(b) (4pts)Supposewe alsoknow th t theplotsinthesamerow willhavethesameamount
ofwaterirrigation;whilethe plotsin the different rows mayhave differentamountof
waterirrigation. Theyieldoftomatomaybe affected by variabilityinthe amountof
availablemoisture. With this information,whichdesign doyou suggest? owshould
theagronomistdotherandomization?
2
8/11/2019 1112SEM1-ST5203
3/14
ST5203
(c) (4pts) Inadditionto theinformationin (b),we also know
that
theplotsinthesame
columnwillhavethesameamountofsunshinejwhiletheplotsinthedifferentcolumns
may have different amount of sun shine. The yield of tomato may be affected by
variabilityin the amount of available sun shine. With this extra information,which
designdo yousuggest? Howshould the agronomistdotherandomization?
2.
(12pts) The Kenton Companywished
to
test 4different packagedesigns for a new break
fast cereal. Sixteen stores, with approximately equal sales volumes, were selected as the
experimentalunits. Eachstorewasrandomlyassignedoneof thepackagedesigns,witheach
packagedesign assigned
to
4stores. Sales, innumber ofcases, were observedfor the study
period,
and
theresultsarerecordedinTable
1.
The
data
isanalyzedusingone-waylayout.
Packagedesign
Sales Samplemean yd
E l
y
..
_y 2
I j l 1 1
1 11 17 16
14
14.5
21
2
12
1
15 19
14.0
46
3
23 20 18
17
19.5
21
4
27
33
22
26 27.0
62
3
8/11/2019 1112SEM1-ST5203
4/14
ST5203
(a) (3pts) Fill up the blanks in the NOV table below.
Resource Degrees of freedom Sum of squres Mean of squares
Treatment
Error
(b) (3pts) Conduct an F test to test the null hypothesis
th t
the four package designs have
the same sales. Note:
O 05
,3,12 =
3.49.
4
8/11/2019 1112SEM1-ST5203
5/14
ST5203
(c) (3pts) Use the Thkey method to perform multiple comparisons of the four package
designs
t
the 0.05 leveL Note
QO.05,4,12=4.20.
(d) (3pts) Suppose the interest
is
to test the null hypothesis
th t
the mean sales
for
design
1
is the same as the average sales of all four designs. Conduct a t test to this null
hypothesis
t a
0.05. Note:
to.0
25
12 2.18.
5
http:///reader/full/QO.05,4,12=4.20http:///reader/full/QO.05,4,12=4.208/11/2019 1112SEM1-ST5203
6/14
ST5203
3. (12pts)A company conducted a
27
full factorial experiment with factors
A B
C, D
E
F G.
Then, the
data
is processed
by
Yate's algorithm in R and all the computed effects,
including
the estimated grand
mean,
are stored in the txt file "Effects_output.txt" in standard
order. Assume that the computations in Yate's algorithm are all correct. Read the following
R outputs to answer all the sub-questions.
>
#
Load data,
> "Effects_output.txt" contains estimates
of effects
from Yate's algorithm
>
Effects_data =
read.table(IEffects_output.txt", header
= F)
> Effects
=
Effects_data$Vl
>
#
Remove grandmean
>
Effects = Effects[2:length(Effects)]
>
#
Order
the
effects
> Effects a
= abs(Effects)
> Effects_a_order = order(Effects_a)
>
#
Number
of
effects
> I
= length(Effects)
> # Some
possibly
relevant
information
> cbind(Effects_a_order[(I-8):I], Effects_a[Effects_a_order[(I-8):I]])
[,1] [,2]
[1,] 97 2.043181
[2,]
113
2.119931
[3,]
16 2.434177
[4,]
111 2.456506
[5,]
116 2.800000
[6,]
114
3.000000
[7,]
85
12.000000
[8,]
31
15.000000
[9,]
11
20.000000
6
8/11/2019 1112SEM1-ST5203
7/14
ST5203
>
# Some possibly relevant
computational
results
median Effects_a)
[ ] 0.7903197
median Effects_a[Effects_a
8/11/2019 1112SEM1-ST5203
8/14
ST5203
(c)
3pts)Six students attending ST5203 independently constructed half normal plots based
upon the computed effects. Only one student is doing the correct work, which one is
it? Point
out
at least one mistake for each of the other plots.
0
N
'
,
;::
1 l
::J
J:i
..
'
0
0
N
'
II
-
,
1!l
-
::J
J:i
'
0
Student 1
0
0
0
o
0
I
2
1
0
2
half
normalquantile
Student3
0
0
0
. cs;tPO
00
-
I I I
0.0
O.S
1,0
1.S 2.0 2.S
halfnormalquantile
0
N
Student2
0
1 l
::J
5
III
J:i
IV
'
-
0
-
'
0
..-
0,0
0.5
0
0
~
1,0 1,S 2,0
2.S
halfnormalquantile
0
N
Student4
0
'"
..
"0
III
J:i
IV
';!
;::
'
0
-
I
0.0 0.5
0
0
~
I I
1.0 1.5 2.0 2.5
half
normalquantile
8
8/11/2019 1112SEM1-ST5203
9/14
ST5203
Student
Student 6
0
N
It
5
0
III
5
co
.0
\0
'
a
0
0
0
.
~
0
N
0
0
5
rg
0
CD
a
:::J
5
'
0
'
\0
a
0
~
0.0
0.5
1.0
1.5
2.0
2.5
2
1
0
2
half normal quantile half normal quantile
(d) (3pts)With the correct half-normal plot in (c),
how
many effects should be declared as
significant? Clearly write down the name{s) of your declared significant effects.
9
8/11/2019 1112SEM1-ST5203
10/14
ST5203
4
(12pts}Assume
th t five
factors
A, B, C, D
and
E),
each
t
two levels, are studied with
runs given by the following treatment combinations (I),
ade
bd abe
ede
ae
bee
and
abed.
(a) (3pts)Write down the design matrix of the experiment.
(b) (3pts}Write down the defining contrast subgroup of the design.
1
8/11/2019 1112SEM1-ST5203
11/14
ST5203
(c)
{3pts)Are there any clear effects in this design? yes, list all clear effects. no, explain
why
not.
(d) {3pts)Denote the above design as dll consider another 8-run design d
2
by switching the
signs in all columns of d
1
Consider d
3
to the augmented design of d
1
and d
2
i.e., d
3
has 16 rows. Find a set of generators
for
d
3
hat
has been achieved by adding d
2
?
Explain clearly.
11
8/11/2019 1112SEM1-ST5203
12/14
ST5203
5. (12pts)Replicate2
3
fullfactorialexperimentswithfactorsA
Band
C3times. Eachreplica
tion
is
arrangedin4blocks. Theeffectsfor constructingtheblockingschemes for replicates
aredisplayedasfollows.
Replicate
1: Dl AC
D
ABC;
Replicate2:
I=BC D =ABC;
Replicate3: Dl AB D =
AC.
After collectingthe data from the experiment, Yate's algorithm is applied to each ofthe
replicates.The obtainedeffect estimatesfromeveryreplicatearedisplayedin the following
tablein
st nd rd order
Rep. 1 Rep. 2 Rep. 3
2.76
3.94 0.59
-0.30 3.36 3.34
8.46
4.96 4.93
-4.09 -1.92 -1.04
2.16 -1.65 -2.29
4.71
4.91
6.17
-0.38 0.64 0.48
-0.26 1.41 2.32
12
8/11/2019 1112SEM1-ST5203
13/14
ST5203
(a) (3pts)Obtain the estimates
for
ll the effects in this experiment.
(b) (3pts)Estimate j2, the variance for each individual observation. hat are the degrees
of freedom of your estimates.
13
8/11/2019 1112SEM1-ST5203
14/14
ST5203
(c) 3pts)Construct
95
confidence intervals
for
effects A and G Note: to 0
25
5 2.57.
d) 3pts)Construct 95 confidence intervals for effects
AB
and
BG
nd of
Paper