A little about me
自己紹介
- オーストラリア育ちの日本語がイマイチなJapanese-Australianです🇯🇵🇦🇺
- ５年前に統計学の博士号を取得しました🎓
- Monash大学で働くため２０２０年１月シドニーからメルボルンへ引っ越しました📦
私のオフィスからの見晴らしです
1
Grammar of Graphics
Wickham (2016) ggplot2: Elegant Graphics for Data Analysis. Springer-Verlag New York
Wilkinson (2005) The Grammar of Graphics. Statistics and Computing. Springer, 2nd edition.
Base R
df## duty perc## 1 Teaching 40## 2 Research 40## 3 Other 20tibble::as_tibble(df)## # A tibble: 3 x 2## duty perc## <chr> <dbl>## 1 Teaching 40## 2 Research 40## 3 Other 20barplot(as.matrix(df$perc), legend = df$duty)
pie(df$perc, labels = df$duty)
R Core Team (2020) R: A Language and Environment for Statistical Computing https://www.R-project.org/
Base R
df## duty perc## 1 Teaching 40## 2 Research 40## 3 Other 20tibble::as_tibble(df)## # A tibble: 3 x 2## duty perc## <chr> <dbl>## 1 Teaching 40## 2 Research 40## 3 Other 20barplot(as.matrix(df$perc), legend = df$duty)
pie(df$perc, labels = df$duty)
R Core Team (2020) R: A Language and Environment for Statistical Computing https://www.R-project.org/
Single purpose functions to generate "named plots"
ggplot2
df## duty perc## 1 Teaching 40## 2 Research 40## 3 Other 20tibble::as_tibble(df)## # A tibble: 3 x 2## duty perc## <chr> <dbl>## 1 Teaching 40## 2 Research 40## 3 Other 20ggplot(df, aes(x = "", # dummy y = perc, fill = duty)) + geom_col()
ggplot2
df## duty perc## 1 Teaching 40## 2 Research 40## 3 Other 20tibble::as_tibble(df)## # A tibble: 3 x 2## duty perc## <chr> <dbl>## 1 Teaching 40## 2 Research 40## 3 Other 20ggplot(df, aes(x = "", # dummy y = perc, fill = duty)) + geom_col()
ggplot(df, aes(x = "", # dummy y = perc, fill = duty)) + geom_col() + coord_polar(theta = "y")
ggplot2
df## duty perc## 1 Teaching 40## 2 Research 40## 3 Other 20tibble::as_tibble(df)## # A tibble: 3 x 2## duty perc## <chr> <dbl>## 1 Teaching 40## 2 Research 40## 3 Other 20ggplot(df, aes(x = "", # dummy y = perc, fill = duty)) + geom_col()
ggplot(df, aes(x = "", # dummy y = perc, fill = duty)) + geom_col() + coord_polar(theta = "y")
stacked barplotとpie chartの違いはcoordinate systemがCartesian coordinateからpolar coordinateと変わった
ポテトフィールド研究・このプロットはどうコードする？
data(cochran.crd, package = "agridat")cochran.crd## inf trt row col## 1 9 F3 4 1## 2 12 O 4 2## 3 18 S6 4 3## 4 10 F12 4 4## 5 24 S6 4 5## 6 17 S12 4 6## 7 30 S3 4 7## 8 16 F6 4 8## 9 10 O 3 1## 10 7 S3 3 2## 11 4 F12 3 3## 12 10 F6 3 4## 13 21 S3 3 5## 14 24 O 3 6## 15 29 O 3 7## 16 12 S6 3 8## 17 9 F3 2 1## 18 7 S12 2 2## 19 18 F6 2 3## 20 30 O 2 4## 21 18 F6 2 5## 22 16 S12 2 6## 23 16 F3 2 7## 24 4 F12 2 8## 25 9 S3 1 1## 26 18 O 1 2## 27 17 S12 1 3## 28 19 S6 1 4## 29 32 O 1 5## 30 5 F12 1 6## 31 26 O 1 7## 32 4 F3 1 8
Kevin Wright (2020). agridat: Agricultural Datasets. R
package version 1.17.
W.G. Cochran and G. Cox, 1957. Experimental Designs, 2nd ed. John Wiley, New York.
ポテトフィールド研究・このプロットはどうコードする？
data(cochran.crd, package = "agridat")cochran.crd## inf trt row col## 1 9 F3 4 1## 2 12 O 4 2## 3 18 S6 4 3## 4 10 F12 4 4## 5 24 S6 4 5## 6 17 S12 4 6## 7 30 S3 4 7## 8 16 F6 4 8## 9 10 O 3 1## 10 7 S3 3 2## 11 4 F12 3 3## 12 10 F6 3 4## 13 21 S3 3 5## 14 24 O 3 6## 15 29 O 3 7## 16 12 S6 3 8## 17 9 F3 2 1## 18 7 S12 2 2## 19 18 F6 2 3## 20 30 O 2 4## 21 18 F6 2 5## 22 16 S12 2 6## 23 16 F3 2 7## 24 4 F12 2 8## 25 9 S3 1 1## 26 18 O 1 2## 27 17 S12 1 3## 28 19 S6 1 4## 29 32 O 1 5## 30 5 F12 1 6## 31 26 O 1 7## 32 4 F3 1 8
ggplot(cochran.crd, aes(x = col, y = row, fill = inf)) + geom_tile(color = "black", size = 1.2) + scale_fill_gradient(low = "gold", high = "firebrick4") + theme(text = element_text(size = 20))
ggplot(cochran.crd, aes(x = col, y = row, fill = trt)) + geom_tile(color = "black", size = 1.2) + scale_fill_viridis_d() + theme(text = element_text(size = 20)) Kevin Wright (2020). agridat: Agricultural Datasets. R
package version 1.17.
W.G. Cochran and G. Cox, 1957. Experimental Designs, 2nd ed. John Wiley, New York.
2
Grammar of
Data Manipulation
Wickham, François, Henry & Müller (2020) dplyr: A Grammar of Data Manipulation. R-package version 1.0.0.
Completely Randomised Design (CRD)
- t treatments randomised to n units
observation=mean+treatment+error
(with constraints and distributional assumptions)
Randomised Complete Block Design (RCBD)
- b blocks of size t
- t treatments randomised to t units within each block
observation=mean+treatment+block+error
Latin Square Design (LSD)
- two orthogonal blocks of size t
- t treatments randomised to units such that every treatment appears exactly once in each block
observation=mean+treatment+row+column+error
Balanced Incomplete Block Design (BIBD)
- b blocks of size k<t
- t treatments randomised to units within each block such that every pair of treatment appears the same number of times across blocks
observation=mean+block+treatment+error
Factorial Design
- ab=t treatments randomised to n units
- treatment is every combination of two factors A and B
observation=mean+A+B+A:B+error
Split-plot Design
- n1 whole plots consisting of b sub plots
- in total there are n sub plots
- treatment factor A is randomised to whole plots
- treatment factor B is randomised to sub plots within each whole plot
observation=mean+A+WP+B+A:B+error
ダウンロードされたR-packagesトップ１０
AlgDesign and agricolae are the most downloaded
(data from cranlogs from 2015-01-01 to 2020-12-31)
agricolae::design.crd
Completely randomised design for t=3 treatments with 2 replicates each
trt <- c("A", "B", "C")agricolae::design.crd(trt = trt, r = 2) %>% glimpse()## List of 2## $ parameters:List of 7## ..$ design: chr "crd"## ..$ trt : chr [1:3] "A" "B" "C"## ..$ r : num [1:3] 2 2 2## ..$ serie : num 2## ..$ seed : int 2074805913## ..$ kinds : chr "Super-Duper"## ..$ : logi TRUE## $ book :'data.frame': 6 obs. of 3 variables:## ..$ plots: num [1:6] 101 102 103 104 105 106## ..$ r : int [1:6] 1 2 1 1 2 2## ..$ trt : chr [1:6] "A" "A" "B" "C" ...
agricolae::design.rcbd
Randomised complete block design for t=3 treatments with 2 blocks
trt <- c("A", "B", "C")agricolae::design.rcbd(trt = trt, r = 2) %>% glimpse()## List of 3## $ parameters:List of 7## ..$ design: chr "rcbd"## ..$ trt : chr [1:3] "A" "B" "C"## ..$ r : num 2## ..$ serie : num 2## ..$ seed : int 490734433## ..$ kinds : chr "Super-Duper"## ..$ : logi TRUE## $ sketch : chr [1:2, 1:3] "A" "C" "B" "A" ...## $ book :'data.frame': 6 obs. of 3 variables:## ..$ plots: num [1:6] 101 102 103 201 202 203## ..$ block: Factor w/ 2 levels "1","2": 1 1 1 2 2 2## ..$ trt : Factor w/ 3 levels "A","B","C": 1 2 3 3 1 2
agricolae::design.lsd()
Latin square design for t=3 treatments
trt <- c("A", "B", "C")agricolae::design.lsd(trt = trt) %>% glimpse()## List of 3## $ parameters:List of 7## ..$ design: chr "lsd"## ..$ trt : chr [1:3] "A" "B" "C"## ..$ r : int 3## ..$ serie : num 2## ..$ seed : int -338489375## ..$ kinds : chr "Super-Duper"## ..$ : logi TRUE## $ sketch : chr [1:3, 1:3] "A" "C" "B" "B" ...## $ book :'data.frame': 9 obs. of 4 variables:## ..$ plots: num [1:9] 101 102 103 201 202 203 301 302 303## ..$ row : Factor w/ 3 levels "1","2","3": 1 1 1 2 2 2 3 3 3## ..$ col : Factor w/ 3 levels "1","2","3": 1 2 3 1 2 3 1 2 3## ..$ trt : Factor w/ 3 levels "A","B","C": 1 2 3 3 1 2 2 3 1
agricolae::design.bib()
Balanced incomplete block design for t=3 treatments with block size of 2
trt <- c("A", "B", "C")agricolae::design.bib(trt = trt, k = 2) %>% glimpse()## [1] "No improvement over initial random design."## ## Parameters BIB## ==============## Lambda : 1## treatmeans : 3## Block size : 2## Blocks : 3## Replication: 2 ## ## Efficiency factor 0.75 ## ## <<< Book >>>## List of 4## $ parameters:List of 6## ..$ design: chr "bib"## ..$ trt : chr [1:3] "A" "B" "C"## ..$ k : num 2## ..$ serie : num 2## ..$ seed : int -907106719## ..$ kinds : chr "Super-Duper"## $ statistics:'data.frame': 1 obs. of 6 variables:## ..$ lambda : num 1## ..$ treatmeans: int 3## ..$ blockSize : num 2## ..$ blocks : int 3## ..$ r : num 2## ..$ Efficiency: num 0.75## $ sketch : chr [1:3, 1:2] "C" "A" "C" "A" ...## $ book :'data.frame': 6 obs. of 3 variables:## ..$ plots: num [1:6] 101 102 201 202 301 302## ..$ block: Factor w/ 3 levels "1","2","3": 1 1 2 2 3 3## ..$ trt : Factor w/ 3 levels "A","B","C": 3 1 1 2 3 2
agricolae::design.ab()
Factorial design for t=3×2 treatments with 2 replication for each treatment
agricolae::design.ab(trt = c(3, 2), r = 2, design = "crd") %>% glimpse()## List of 2## $ parameters:List of 8## ..$ design : chr "factorial"## ..$ trt : chr [1:6] "1 1" "1 2" "2 1" "2 2" ...## ..$ r : num [1:6] 2 2 2 2 2 2## ..$ serie : num 2## ..$ seed : int -751356955## ..$ kinds : chr "Super-Duper"## ..$ : logi TRUE## ..$ applied: chr "crd"## $ book :'data.frame': 12 obs. of 4 variables:## ..$ plots: num [1:12] 101 102 103 104 105 106 107 108 109 110 ...## ..$ r : int [1:12] 1 1 1 1 1 2 2 1 2 2 ...## ..$ A : chr [1:12] "3" "2" "3" "1" ...## ..$ B : chr [1:12] "1" "2" "2" "2" ...Note not A/B testing!
agricolae::design.split()
Split-plot design for t=2×4 treatments with 2 replication for each treatment
trt1 <- c("I", "R"); trt2 <- LETTERS[1:4]agricolae::design.split(trt1 = trt1, trt2 = trt2, r = 2, design = "crd") %>% glimpse()## List of 2## $ parameters:List of 8## ..$ design : chr "split"## ..$ : logi TRUE## ..$ trt1 : chr [1:2] "I" "R"## ..$ applied: chr "crd"## ..$ r : num [1:2] 2 2## ..$ serie : num 2## ..$ seed : int 157974485## ..$ kinds : chr "Super-Duper"## $ book :'data.frame': 16 obs. of 5 variables:## ..$ plots : num [1:16] 101 101 101 101 102 102 102 102 103 103 ...## ..$ splots: Factor w/ 4 levels "1","2","3","4": 1 2 3 4 1 2 3 4 1 2 ...## ..$ r : int [1:16] 1 1 1 1 1 1 1 1 2 2 ...## ..$ trt1 : chr [1:16] "R" "R" "R" "R" ...## ..$ trt2 : chr [1:16] "C" "B" "A" "D" ...
実験計画法の文法とは
- Base R プロット覚えてますか？ base plot
- 🤔 "named experimental design" functions (
agricolae::design.crd, etc.) are like "named statistical graphic" functions (pie,barplot) - 💡 このコンセプトが
edibbleR-packageのもとです
https://github.com/emitanaka/edibble
名前の元
tibbleR-package とは modern reimagining of thedata.frameedibbleとは experimental design tibble
Müller & Wickham (2020) tibble: Simple Data Frames. R package version 3.0.3
