planor-package: Generation of Regular Factorial Designs
In planor: Generation of Regular Factorial Designs
Description
Details
Note
Author(s)
References
Examples
Description
A package dedicated to the automatic generation of regular
factorial designs, including fractional designs, orthogonal block designs,
row-column designs and split-plots.
Details
The user describes the factors to be controlled in the
experiment and the anova model to be used when the results will be
analysed. He or she also specifies the size of the design, that is,
the number of available experimental units. Then planor looks for a
design satisfying these specifications and possibly randomizes it. The
core of the algorithm is the search for the key matrix, an integer
matrix which determines the aliasing in the resulting factorial
design.
The user may use the function regular.design where all
these steps are integrated, and transparent by default. Alternatively,
the steps can be decomposed by using successively the functions
planor.factors, planor.model,
planor.designkey and
planor.design. For the expert
user, the function planor.designkey can give several key
matrix solutions. Alias and summary methods allow to study and compare
these solutions, in order to select the most appropriate one for the
final design.
Note
An R option named planor.max.print is set. It is equal
to the number of printed
rows and columns in the display of planor matrices. Default is 20.
You can change its value by using the function
options() (see ?options).
Author(s)
Monod, H. herve.monod@inrae.fr and Bouvier, A. and Kobilinsky, A.
References
See citation("planor").
Examples
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
# DESIGN SPECIFICATIONS
# Treatments: four 3-level factors A, B, C, D
# Units: 27 in 3 blocks of size 9
# Non-negligible factorial terms:
# block + A + B + C + D + A:B + A:C + A:D + B:C + B:D + C:D
# Factorial terms to estimate:
# A + B + C + D
# 1. DIRECT GENERATION, USING regular.design
mydesign <- regular.design(factors=c("block", LETTERS[1:4]),
nlevels=rep(3,5), model=~block+(A+B+C+D)^2, estimate=~A+B+C+D,
nunits=3^3, randomize=~block/UNITS)
print(mydesign)
# DUMMY ANALYSIS
# Here we omit two-factor interactions from the model, so they are
# confounded with the residuals (but not with ABCD main effects)
set.seed(123)
mydesigndata <- mydesign@design
mydesigndata$Y <- runif(27)
mydesign.aov <- aov(Y ~ block + A + B + C + D, data=mydesigndata)
summary(mydesign.aov)
# 2. STEP-BY-STEP GENERATION, USING planor.designkey
F0 <- planor.factors(factors=c( "block", LETTERS[1:4]), nlevels=rep(3,5),
block=~block)
M0 <- planor.model(model=~block+(A+B+C+D)^2, estimate=~A+B+C+D)
K0 <- planor.designkey(factors=F0, model=M0, nunits=3^3, max.sol=2)
summary(K0)
mydesign.S4 <- planor.design(key=K0, select=2)
Example output
Loaded planor 1.3.7
The search is closed: max.sol = 1 solution(s) found
An object of class "planordesign"
Slot "design":
block A B C D
1 1 3 2 3 2
2 1 3 2 2 1
3 1 1 3 1 3
4 1 1 3 3 2
5 1 1 3 2 1
6 1 2 1 3 2
7 1 2 1 2 1
8 1 3 2 1 3
9 1 2 1 1 3
10 2 1 1 1 1
11 2 3 3 3 3
12 2 2 2 2 2
13 2 3 3 1 1
14 2 3 3 2 2
15 2 2 2 3 3
16 2 1 1 3 3
17 2 1 1 2 2
18 2 2 2 1 1
19 3 2 3 3 1
20 3 3 1 3 1
21 3 2 3 2 3
22 3 3 1 1 2
23 3 3 1 2 3
24 3 1 2 1 2
25 3 2 3 1 2
26 3 1 2 2 3
27 3 1 2 3 1
Slot "factors":
An object of class "designfactors"
Slot "fact.info":
nlev block ordered model basic dummy
block 3 FALSE FALSE TRUE FALSE FALSE
A 3 FALSE FALSE TRUE FALSE FALSE
B 3 FALSE FALSE TRUE FALSE FALSE
C 3 FALSE FALSE TRUE FALSE FALSE
D 3 FALSE FALSE TRUE FALSE FALSE
Slot "pseudo.info":
parent nlev block ordered model basic dummy
block 1 3 FALSE FALSE TRUE FALSE FALSE
A 2 3 FALSE FALSE TRUE FALSE FALSE
B 3 3 FALSE FALSE TRUE FALSE FALSE
C 4 3 FALSE FALSE TRUE FALSE FALSE
D 5 3 FALSE FALSE TRUE FALSE FALSE
Slot "levels":
$block
[1] 1 2 3
$A
[1] 1 2 3
$B
[1] 1 2 3
$C
[1] 1 2 3
$D
[1] 1 2 3
Slot "model":
list()
Slot "designkey":
[[1]]
An object of class keymatrix
********** Prime 3 design **********
block A B C D
*U* 1 0 1 0 1
*U* 0 1 1 0 0
*U* 0 0 0 1 1
Slot "nunits":
[1] 27
Slot "recursive":
[1] FALSE
Df Sum Sq Mean Sq F value Pr(>F)
block 2 0.2227 0.11136 1.540 0.245
A 2 0.5385 0.26924 3.724 0.047 *
B 2 0.1471 0.07354 1.017 0.384
C 2 0.0325 0.01627 0.225 0.801
D 2 0.0851 0.04256 0.589 0.567
Residuals 16 1.1569 0.07230
---
Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
Preliminary step 1 : processing the model specifications
Preliminary step 2 : performing prime decompositions on the factors
*** Main step for prime p = 3 : key-matrix search
=> search for columns 2 to 5
first visit to column 2
first visit to column 3
--- col. 3 ( j = 2) 22 selected candidates
first visit to column 4
--- col. 4 ( j = 3) 18 selected candidates
first visit to column 5
--- col. 5 ( j = 4) 8 selected candidates
The search is closed: max.sol = 2 solution(s) found
********** Prime 3 design **********
--- Solution 1 for prime 3 ---
TREATMENT EFFECTS CONFOUNDED WITH THE MEAN
1 = A B^2 C^2 D
BLOCK-and-TREATMENT EFFECTS CONFOUNDED WITH THE MEAN
1 = block^2 A^2 B
1 = block^2 C^2 D
1 = block A^2 B C^2 D
WEIGHT PROFILES
Treatment effects confounded with the mean: 4^1
Treatment effects confounded with block effects: 2^2 4^1
Treatment pseudo-effects confounded with the mean: 4^1
Treatment pseudo-effects confounded with block effects: 2^2 4^1
--- Solution 2 for prime 3 ---
TREATMENT EFFECTS CONFOUNDED WITH THE MEAN
1 = A^2 B C^2 D
BLOCK-and-TREATMENT EFFECTS CONFOUNDED WITH THE MEAN
1 = block^2 A^2 B
1 = block C^2 D
1 = block^2 A B^2 C^2 D
WEIGHT PROFILES
Treatment effects confounded with the mean: 4^1
Treatment effects confounded with block effects: 2^2 4^1
Treatment pseudo-effects confounded with the mean: 4^1
Treatment pseudo-effects confounded with block effects: 2^2 4^1
planor documentation built on March 19, 2020, 1:06 a.m.
R Package Documentation
Browse R Packages
We want your feedback!
Note that we can't provide technical support on individual packages. You should contact the package authors for that.
Description Details Note Author(s) References Examples
Description
A package dedicated to the automatic generation of regular factorial designs, including fractional designs, orthogonal block designs, row-column designs and split-plots.
Details
The user describes the factors to be controlled in the experiment and the anova model to be used when the results will be analysed. He or she also specifies the size of the design, that is, the number of available experimental units. Then planor looks for a design satisfying these specifications and possibly randomizes it. The core of the algorithm is the search for the key matrix, an integer matrix which determines the aliasing in the resulting factorial design.
The user may use the function regular.design where all
these steps are integrated, and transparent by default. Alternatively,
the steps can be decomposed by using successively the functions
planor.factors, planor.model,
planor.designkey and
planor.design. For the expert
user, the function planor.designkey can give several key
matrix solutions. Alias and summary methods allow to study and compare
these solutions, in order to select the most appropriate one for the
final design.
Note
An R option named planor.max.print is set. It is equal
to the number of printed
rows and columns in the display of planor matrices. Default is 20.
You can change its value by using the function
options() (see ?options).
Author(s)
Monod, H. herve.monod@inrae.fr and Bouvier, A. and Kobilinsky, A.
References
See citation("planor").
Examples
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 | # DESIGN SPECIFICATIONS
# Treatments: four 3-level factors A, B, C, D
# Units: 27 in 3 blocks of size 9
# Non-negligible factorial terms:
# block + A + B + C + D + A:B + A:C + A:D + B:C + B:D + C:D
# Factorial terms to estimate:
# A + B + C + D
# 1. DIRECT GENERATION, USING regular.design
mydesign <- regular.design(factors=c("block", LETTERS[1:4]),
nlevels=rep(3,5), model=~block+(A+B+C+D)^2, estimate=~A+B+C+D,
nunits=3^3, randomize=~block/UNITS)
print(mydesign)
# DUMMY ANALYSIS
# Here we omit two-factor interactions from the model, so they are
# confounded with the residuals (but not with ABCD main effects)
set.seed(123)
mydesigndata <- mydesign@design
mydesigndata$Y <- runif(27)
mydesign.aov <- aov(Y ~ block + A + B + C + D, data=mydesigndata)
summary(mydesign.aov)
# 2. STEP-BY-STEP GENERATION, USING planor.designkey
F0 <- planor.factors(factors=c( "block", LETTERS[1:4]), nlevels=rep(3,5),
block=~block)
M0 <- planor.model(model=~block+(A+B+C+D)^2, estimate=~A+B+C+D)
K0 <- planor.designkey(factors=F0, model=M0, nunits=3^3, max.sol=2)
summary(K0)
mydesign.S4 <- planor.design(key=K0, select=2)
|
Example output
Loaded planor 1.3.7
The search is closed: max.sol = 1 solution(s) found
An object of class "planordesign"
Slot "design":
block A B C D
1 1 3 2 3 2
2 1 3 2 2 1
3 1 1 3 1 3
4 1 1 3 3 2
5 1 1 3 2 1
6 1 2 1 3 2
7 1 2 1 2 1
8 1 3 2 1 3
9 1 2 1 1 3
10 2 1 1 1 1
11 2 3 3 3 3
12 2 2 2 2 2
13 2 3 3 1 1
14 2 3 3 2 2
15 2 2 2 3 3
16 2 1 1 3 3
17 2 1 1 2 2
18 2 2 2 1 1
19 3 2 3 3 1
20 3 3 1 3 1
21 3 2 3 2 3
22 3 3 1 1 2
23 3 3 1 2 3
24 3 1 2 1 2
25 3 2 3 1 2
26 3 1 2 2 3
27 3 1 2 3 1
Slot "factors":
An object of class "designfactors"
Slot "fact.info":
nlev block ordered model basic dummy
block 3 FALSE FALSE TRUE FALSE FALSE
A 3 FALSE FALSE TRUE FALSE FALSE
B 3 FALSE FALSE TRUE FALSE FALSE
C 3 FALSE FALSE TRUE FALSE FALSE
D 3 FALSE FALSE TRUE FALSE FALSE
Slot "pseudo.info":
parent nlev block ordered model basic dummy
block 1 3 FALSE FALSE TRUE FALSE FALSE
A 2 3 FALSE FALSE TRUE FALSE FALSE
B 3 3 FALSE FALSE TRUE FALSE FALSE
C 4 3 FALSE FALSE TRUE FALSE FALSE
D 5 3 FALSE FALSE TRUE FALSE FALSE
Slot "levels":
$block
[1] 1 2 3
$A
[1] 1 2 3
$B
[1] 1 2 3
$C
[1] 1 2 3
$D
[1] 1 2 3
Slot "model":
list()
Slot "designkey":
[[1]]
An object of class keymatrix
********** Prime 3 design **********
block A B C D
*U* 1 0 1 0 1
*U* 0 1 1 0 0
*U* 0 0 0 1 1
Slot "nunits":
[1] 27
Slot "recursive":
[1] FALSE
Df Sum Sq Mean Sq F value Pr(>F)
block 2 0.2227 0.11136 1.540 0.245
A 2 0.5385 0.26924 3.724 0.047 *
B 2 0.1471 0.07354 1.017 0.384
C 2 0.0325 0.01627 0.225 0.801
D 2 0.0851 0.04256 0.589 0.567
Residuals 16 1.1569 0.07230
---
Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
Preliminary step 1 : processing the model specifications
Preliminary step 2 : performing prime decompositions on the factors
*** Main step for prime p = 3 : key-matrix search
=> search for columns 2 to 5
first visit to column 2
first visit to column 3
--- col. 3 ( j = 2) 22 selected candidates
first visit to column 4
--- col. 4 ( j = 3) 18 selected candidates
first visit to column 5
--- col. 5 ( j = 4) 8 selected candidates
The search is closed: max.sol = 2 solution(s) found
********** Prime 3 design **********
--- Solution 1 for prime 3 ---
TREATMENT EFFECTS CONFOUNDED WITH THE MEAN
1 = A B^2 C^2 D
BLOCK-and-TREATMENT EFFECTS CONFOUNDED WITH THE MEAN
1 = block^2 A^2 B
1 = block^2 C^2 D
1 = block A^2 B C^2 D
WEIGHT PROFILES
Treatment effects confounded with the mean: 4^1
Treatment effects confounded with block effects: 2^2 4^1
Treatment pseudo-effects confounded with the mean: 4^1
Treatment pseudo-effects confounded with block effects: 2^2 4^1
--- Solution 2 for prime 3 ---
TREATMENT EFFECTS CONFOUNDED WITH THE MEAN
1 = A^2 B C^2 D
BLOCK-and-TREATMENT EFFECTS CONFOUNDED WITH THE MEAN
1 = block^2 A^2 B
1 = block C^2 D
1 = block^2 A B^2 C^2 D
WEIGHT PROFILES
Treatment effects confounded with the mean: 4^1
Treatment effects confounded with block effects: 2^2 4^1
Treatment pseudo-effects confounded with the mean: 4^1
Treatment pseudo-effects confounded with block effects: 2^2 4^1
R Package Documentation
Browse R Packages
We want your feedback!
Note that we can't provide technical support on individual packages. You should contact the package authors for that.
Embedding an R snippet on your website
Add the following code to your website.
For more information on customizing the embed code, read Embedding Snippets.