Sunteți pe pagina 1din 5

Chapter 15: Factorial Data

Section 15.2: The Two-Factor Factorial


15.2.1 (a)   7, 1  2,  2  2, 1  0,  2  0,  11  1,  12  1,  21  1,
 22  1
(b) no

2 2 5 5
15.2.3   8, 1  5,  2  1,  3  4, 1   ,  2  ,  11  ,  12   ,
3 3 3 3
1 1 4 4
 21   ,  22  ,  31   ,  32  ; A and B are not additive for any
3 3 3 3
values of A and B.
r r
L r n
L
15.2.5 Since   0    ij ,  2 Yijk      j    1 . Setting  0 shows that
 j  j
i
i 1 i 1 i 1 k 1
r n

 Y
i 1 k 1
ijk
  Y . j .  Y
 T. j.  rn  rn j , so  j

15.2.7 (a) A = 5, B = 2
2
(b) 2 Y221k is larger
k 1

15.2.9 Yes, Equation 15.2.3 assumes that the underlying within-cell error variances are all equal. For
these data, though, the sample within-cell error variances are considerably different.

Section 15.3: Sums of Squares for Two-Factor Factorials


15.3.3 (a)
Source df SS MS
A 2 4.75 2.38
B 3 30.46 10.15
AB 6 9.92 1.65
Error 12 12.50 1.04
Total 23 57.63

  1.875, 
(b) ˆ1  0.25, ˆ 2  0.625, ˆ 3  0.375,    0.792, 
  0.958,
1 2 3

  0.125
4

   ij
B1 B2 B3 B4
A1 –1.25 0.42 0.58 0.25
A2 0.38 –0.46 –0.30 0.38
A3 0.88 0.05 –0.30 –0.63
(c) No

Copyright © 2018 Pearson Education, Inc.


131
132 Chapter 15: Factorial Data

15.3.5 Yes, because the main events and interactions are all independent, so knowing the numerical
value of one implies nothing about the numerical value of another.

15.3.7
Source df SS MS
A 5 60 12
B 4 44 11
AB 20 18 0.9
Error 60 42
Total 89 164

Section 15.4: Expected Mean Squares


15.4.1 For the model Yij     j   i ( j ) , where A is a fixed effect, the EMS matrix takes the form

k n
F R
j i EMS
2
j 0 n  2  n *
 j (i ) 0 1  2

so the “error” term is the proper denomination for an F test of the A effect.

15.4.3 Below is the data’s ANOVA constructed using the computing formulas on p. 588. The EMS
column is derived from the rules given on p. 15-28.
Source df SS MS EMS
District (Ai) 4 64.45 16.11  2  4 A2
Error 15 15.66 1.04
Total 19 80.11  2

Setting the mean squares equal to their expected values gives ˆ 2  1.04 and ˆ A2  3.77.

Copyright © 2018 Pearson Education, Inc.


Section 15.6: The Three-Factor Factorial Design 133

15.4.5 (a)
r c µ n
F F F R
i j k l
i 0 c µ n  *2
j r 0 µ n  *2
 ij 0 0 µ n
 
*2

rk r c 0 n  r*2
 r ik 0 c 0 n  *2r
  r  jk r 0 0 n
 *2r
 r ijk 0 0 0 n
 
*2
r

 l (ijk ) 1 1 1 1  2

EMS(A)   2  c n *2
EMS(B)   2  r  n *2
EMS  A  B    2   n 
*2

EMS(C)   2  rc r*2


EMS  A  C    2  cn *2r
EMS  B  C    2  rn *2r
EMS  A  B  C    2  n 
*2
r

EMS(error)   2
(b) The appropriate denominator for testing each of the main effects or the interactions is
EMS (error).

Section 15.6: The Three-Factor Factorial Design

 
r c u n 2
15.6.1 L   Yijkl     i   j   k   ij   ik     jk   ijk 
i 1 j 1 k 1 l 1
 
L
 
r c u n
 2 Yijkl     i   j   k   ij   ik     jk   ijk  ( 1)
 i 1 j 1 k 1 l 1
 

L
Setting  0 reduces to

r c u n

 Y
i 1 j 1 k 1 l 1
ijkl  rcun  0

so
1 r c u l
ˆ   Yijlk  Y ....
rcun i 1 j 1 k 1 l 1
The MLE’s for the parameters are derived similarly.

Copyright © 2018 Pearson Education, Inc.


134 Chapter 15: Factorial Data

15.6.3 (a) No, except that all three response pattern are parallel.


(b)   
 Y 111.  Y ....  Y 1...  Y .1..  Y ..1.  Y 11..  Y 1.1.  Y .11.
111

66 30 39 12 15 8 10
6       0
12 6 6 4 3 2 2
 

 112
 Y 112.  Y ....  Y 1...  Y .1..  Y ..2.  Y 11..  Y 1.2.  Y .12.
   

66 30 33 26 15 10 9
 3       0
12 6 6 4 3 2 2
 
If   111

 0 and   112

 0, then    
 0. But if those three are 0, the remaining
113

interactions must also be 0 because of summation constraints. Yes, these data do illustrate the
comment on p. 15-43.

15.6.5 (a) SSA = 8.334; SSB = 16.334; SSC = 45.5; SSD = 8.334; SSAB = 0; SSAC = 3.166;
SSBC = 6.166
(b) SSE = 78; 24
2 2 3 2 T2
(c) SSABCD   ijkl .  C  SSA  SSB  SSC  SSD  SSAB  SSAC  SSAD 
i 1 j 1 k 1 l 1 2

SSBC  SSBD  SSCD  SSABC  SSABD  SSACD  SSBCD

Section 15.7: 2n Designs

15.7.1 CA  3; CB  5; C C  5 ; C AB  3; C


 AC  1; C  ABC  1 .
 BC  3; C
2 2 2 2
None of the factors would be considered additive in the sense that none of the contrasts for
AB, AC, or BC is 0. However, if the analysis of variance was done on these data, the error
term would be formed by pooling the AC and ABC interactions, which, in effect, is treating
those two interactions as though they were 0.

15.7.3
obs. B D BD
(1) 11 – – +
a 5 – – +
b 9 + – –
ab 19 + – –
c 4 – – +
ac 2 – – +
bc 0 + – –
abc 6 + – –
d 3 – + –
ad 16 – + –
bd 14 + + +
abd 20 + + +
cd 8 – + –
acd 7 – + –
bcd 1 + + +
abcd 12 + + +

Copyright © 2018 Pearson Education, Inc.


Section 15.8: Fractional Factorials 135

C BD   1  11  5  9  19  4  2  0  6  3  16  14  20  8  7  1  12
 
8
1 1
   1  ,
8
  8
so factors B and D are not additive.

15.7.5
Effect df SS
A 1 4.5
B 1 12.5
AB 1 18
C 1 12.5
AC 1 2
BC 1 18
ABC 1 0.5
Total 7 68
A reasonable error term would be formed by pooling the AC and ABC interactions. It would
have 2 df and a mean square equal to
 2  0.5  1.25.
2

Section 15.8: Fractional Factorials


15.8.1 The principle 1/2−replicate has 0 or an even number of letters in common with ABC
Treat. A B AB D AD BD ABD
Comb. (BC) (AC) (C) (ABCD) (BCD) (ACD) (CD)
(1) −+ −+ +− −+ +− +− −+
ab +− +− +− −+ −+ −+ −+
ac +− −+ −+ −+ −+ +− +−
bc −+ +− −+ −+ +− −+ +−
d −+ −+ +− +− −+ −+ +−
abd +− +− +− +− +− +− +−
acd +− −+ −+ +− +− −+ −+
bcd −+ +− −+ +− −+ +− −+

15.8.3 A linear combination of treatments that does not represent a main effect or interaction should
never be used as a defining contrast. Doing so would eliminate the orthogonality that exists
among the columns in the effect matrix and would bias the estimates of those effects.

15.8.5 (a) (1), bc, abd, acd


(b) (1), ac, abd, bcd

Copyright © 2018 Pearson Education, Inc.

S-ar putea să vă placă și