如何使用 nparcomp 包从 Tukey 的多重比较中提取紧凑的字母显示？
How can I extract the compact letter display from a Tukey's multiple comparison using the nparcomp package?
multcomp 包内置了一个很好的紧凑字母显示功能，但我使用的是非参数多重比较包 "nparcomp"，它似乎没有类似的功能。我注意到有几个包（例如 multcompView 和 rcompanion）具有 CLD 功能，但我不确定如何让我的 nparcomp 摘要与这些工具合作。也许这里有人可以帮助我？以下是 nparcomp Tukey 测试的示例摘要：
library(nparcomp)
pristineraw.tukey <- mctp(positif.prop.total ~ dose.log, data = pristineraw, type = "Tukey", conf.level = 0.95, asy.method = "fisher", info = FALSE)
pristineraw.tukey
$Data.Info
Sample Size Effect Lower Upper
1 -4 4 0.7812500 0.65095081 0.8724403
2 -2.95860731484178 2 0.8229167 0.68706660 0.9077133
3 -1.99567862621736 4 0.6145833 0.49050216 0.7253656
4 -0.999565922520681 4 0.4166667 0.33069961 0.5080188
5 4.34272768626649e-05 4 0.1562500 0.08581288 0.2675807
6 1.0000043429231 2 0.2083333 0.12491776 0.3266579
$Contrast
1 2 3 4 5 6
2 - 1 -1 1 0 0 0 0
3 - 1 -1 0 1 0 0 0
4 - 1 -1 0 0 1 0 0
5 - 1 -1 0 0 0 1 0
6 - 1 -1 0 0 0 0 1
3 - 2 0 -1 1 0 0 0
4 - 2 0 -1 0 1 0 0
5 - 2 0 -1 0 0 1 0
6 - 2 0 -1 0 0 0 1
4 - 3 0 0 -1 1 0 0
5 - 3 0 0 -1 0 1 0
6 - 3 0 0 -1 0 0 1
5 - 4 0 0 0 -1 1 0
6 - 4 0 0 0 -1 0 1
6 - 5 0 0 0 0 -1 1
$Analysis
Estimator Lower Upper Statistic p.Value
2 - 1 0.042 -0.431 0.496 0.343 0.99761714
3 - 1 -0.167 -0.586 0.323 -1.381 0.69088411
4 - 1 -0.365 -0.648 0.007 -4.062 0.05318202
5 - 1 -0.625 -0.867 -0.144 -5.151 0.02076608
6 - 1 -0.573 -0.838 -0.090 -4.801 0.02785983
3 - 2 -0.208 -0.609 0.277 -1.763 0.50620162
4 - 2 -0.406 -0.688 -0.019 -4.320 0.04250191
5 - 2 -0.667 -0.894 -0.164 -5.205 0.02026988
6 - 2 -0.615 -0.866 -0.115 -4.930 0.02523067
4 - 3 -0.198 -0.583 0.260 -1.775 0.50151321
5 - 3 -0.458 -0.746 -0.026 -4.365 0.04027067
6 - 3 -0.406 -0.712 0.028 -3.880 0.06346250
5 - 4 -0.260 -0.561 0.101 -2.997 0.14893258
6 - 4 -0.208 -0.559 0.206 -2.078 0.37679610
6 - 5 0.052 -0.380 0.466 0.476 0.99043710
$Analysis.Inf
Estimator Lower Upper Statistic p.Value
2 - 1 0.04166667 -0.4310816 0.496466660 0.3426000 0.99761714
3 - 1 -0.16666667 -0.5861671 0.323305915 -1.3807046 0.69088411
4 - 1 -0.36458333 -0.6475061 0.006668684 -4.0618961 0.05318202
5 - 1 -0.62500000 -0.8671346 -0.143918870 -5.1509655 0.02076608
6 - 1 -0.57291667 -0.8375693 -0.090485809 -4.8010534 0.02785983
3 - 2 -0.20833333 -0.6088821 0.276867328 -1.7626807 0.50620162
4 - 2 -0.40625000 -0.6877026 -0.018637527 -4.3195377 0.04250191
5 - 2 -0.66666667 -0.8944430 -0.164222955 -5.2046137 0.02026988
6 - 2 -0.61458333 -0.8659606 -0.115293472 -4.9298694 0.02523067
4 - 3 -0.19791667 -0.5834144 0.260362074 -1.7746828 0.50151321
5 - 3 -0.45833333 -0.7460603 -0.026382368 -4.3654031 0.04027067
6 - 3 -0.40625000 -0.7115636 0.028113118 -3.8797113 0.06346250
5 - 4 -0.26041667 -0.5608547 0.100626889 -2.9973930 0.14893258
6 - 4 -0.20833333 -0.5594223 0.206138515 -2.0776563 0.37679610
6 - 5 0.05208333 -0.3804685 0.465937204 0.4758687 0.99043710
$Overall
Quantile p.Value
1 4.132777 0.02026988
$input
$input$formula
positif.prop.total ~ dose.log
$input$data
dose positif negatif dead totalNb positif.prop.total dose.log
1 0e+00 17 20 0 37 0.45945946 -4.000000e+00
2 0e+00 23 16 0 39 0.58974359 -4.000000e+00
3 0e+00 18 15 0 33 0.54545455 -4.000000e+00
4 0e+00 14 14 1 28 0.50000000 -4.000000e+00
5 1e-03 19 19 1 38 0.50000000 -2.958607e+00
6 1e-03 20 14 4 34 0.58823529 -2.958607e+00
7 1e-02 22 16 0 38 0.57894737 -1.995679e+00
8 1e-02 18 19 0 37 0.48648649 -1.995679e+00
9 1e-02 15 22 2 37 0.40540541 -1.995679e+00
10 1e-02 11 20 4 31 0.35483871 -1.995679e+00
11 1e-01 12 20 0 32 0.37500000 -9.995659e-01
12 1e-01 12 17 4 29 0.41379310 -9.995659e-01
13 1e-01 8 26 3 34 0.23529412 -9.995659e-01
14 1e-01 5 18 11 23 0.21739130 -9.995659e-01
15 1e+00 3 16 10 19 0.15789474 4.342728e-05
16 1e+00 1 16 5 17 0.05882353 4.342728e-05
17 1e+00 2 24 9 26 0.07692308 4.342728e-05
18 1e+00 7 23 6 30 0.23333333 4.342728e-05
19 1e+01 3 10 8 13 0.23076923 1.000004e+00
20 1e+01 2 20 8 22 0.09090909 1.000004e+00
$input$type
[1] "Tukey"
$input$conf.level
[1] 0.95
$input$alternative
[1] "two.sided"
$input$asy.method
[1] "fisher"
$input$plot.simci
[1] FALSE
$input$control
NULL
$input$info
[1] FALSE
$input$rounds
[1] 3
$input$contrast.matrix
NULL
$input$correlation
[1] FALSE
$input$effect
[1] "unweighted"
$input$const
[1] 0.5875441
$text.Output
[1] "True differences of relative effects are not equal to 0"
$text.output.W
[1] "Global Pseudo Ranks"
$connames
[1] "2 - 1" "3 - 1" "4 - 1" "5 - 1" "6 - 1" "3 - 2" "4 - 2" "5 - 2" "6 - 2"
[10] "4 - 3" "5 - 3" "6 - 3" "5 - 4" "6 - 4" "6 - 5"
$AsyMethod
[1] "Fisher with 5 DF"
attr(,"class")
[1] "mctp"
您可以在包 rcompanion 中使用 cldList。您没有提供可重现的数据，所以我将使用 R 中包含的鸢尾花数据集：
data(iris)
library(rcompanion)
library(nparcomp)
iris.mc <- mctp(Sepal.Length~Species, iris)
comp <- iris.mc$connames
pv <- iris.mc$Analysis$p.Value
cldList(comparison=comp, p.value=pv)
# Group Letter MonoLetter
# 1 2 a a
# 2 3 b b
# 3 1 c c
multcomp 包内置了一个很好的紧凑字母显示功能，但我使用的是非参数多重比较包 "nparcomp"，它似乎没有类似的功能。我注意到有几个包（例如 multcompView 和 rcompanion）具有 CLD 功能，但我不确定如何让我的 nparcomp 摘要与这些工具合作。也许这里有人可以帮助我？以下是 nparcomp Tukey 测试的示例摘要：
library(nparcomp)
pristineraw.tukey <- mctp(positif.prop.total ~ dose.log, data = pristineraw, type = "Tukey", conf.level = 0.95, asy.method = "fisher", info = FALSE)
pristineraw.tukey
$Data.Info
Sample Size Effect Lower Upper
1 -4 4 0.7812500 0.65095081 0.8724403
2 -2.95860731484178 2 0.8229167 0.68706660 0.9077133
3 -1.99567862621736 4 0.6145833 0.49050216 0.7253656
4 -0.999565922520681 4 0.4166667 0.33069961 0.5080188
5 4.34272768626649e-05 4 0.1562500 0.08581288 0.2675807
6 1.0000043429231 2 0.2083333 0.12491776 0.3266579
$Contrast
1 2 3 4 5 6
2 - 1 -1 1 0 0 0 0
3 - 1 -1 0 1 0 0 0
4 - 1 -1 0 0 1 0 0
5 - 1 -1 0 0 0 1 0
6 - 1 -1 0 0 0 0 1
3 - 2 0 -1 1 0 0 0
4 - 2 0 -1 0 1 0 0
5 - 2 0 -1 0 0 1 0
6 - 2 0 -1 0 0 0 1
4 - 3 0 0 -1 1 0 0
5 - 3 0 0 -1 0 1 0
6 - 3 0 0 -1 0 0 1
5 - 4 0 0 0 -1 1 0
6 - 4 0 0 0 -1 0 1
6 - 5 0 0 0 0 -1 1
$Analysis
Estimator Lower Upper Statistic p.Value
2 - 1 0.042 -0.431 0.496 0.343 0.99761714
3 - 1 -0.167 -0.586 0.323 -1.381 0.69088411
4 - 1 -0.365 -0.648 0.007 -4.062 0.05318202
5 - 1 -0.625 -0.867 -0.144 -5.151 0.02076608
6 - 1 -0.573 -0.838 -0.090 -4.801 0.02785983
3 - 2 -0.208 -0.609 0.277 -1.763 0.50620162
4 - 2 -0.406 -0.688 -0.019 -4.320 0.04250191
5 - 2 -0.667 -0.894 -0.164 -5.205 0.02026988
6 - 2 -0.615 -0.866 -0.115 -4.930 0.02523067
4 - 3 -0.198 -0.583 0.260 -1.775 0.50151321
5 - 3 -0.458 -0.746 -0.026 -4.365 0.04027067
6 - 3 -0.406 -0.712 0.028 -3.880 0.06346250
5 - 4 -0.260 -0.561 0.101 -2.997 0.14893258
6 - 4 -0.208 -0.559 0.206 -2.078 0.37679610
6 - 5 0.052 -0.380 0.466 0.476 0.99043710
$Analysis.Inf
Estimator Lower Upper Statistic p.Value
2 - 1 0.04166667 -0.4310816 0.496466660 0.3426000 0.99761714
3 - 1 -0.16666667 -0.5861671 0.323305915 -1.3807046 0.69088411
4 - 1 -0.36458333 -0.6475061 0.006668684 -4.0618961 0.05318202
5 - 1 -0.62500000 -0.8671346 -0.143918870 -5.1509655 0.02076608
6 - 1 -0.57291667 -0.8375693 -0.090485809 -4.8010534 0.02785983
3 - 2 -0.20833333 -0.6088821 0.276867328 -1.7626807 0.50620162
4 - 2 -0.40625000 -0.6877026 -0.018637527 -4.3195377 0.04250191
5 - 2 -0.66666667 -0.8944430 -0.164222955 -5.2046137 0.02026988
6 - 2 -0.61458333 -0.8659606 -0.115293472 -4.9298694 0.02523067
4 - 3 -0.19791667 -0.5834144 0.260362074 -1.7746828 0.50151321
5 - 3 -0.45833333 -0.7460603 -0.026382368 -4.3654031 0.04027067
6 - 3 -0.40625000 -0.7115636 0.028113118 -3.8797113 0.06346250
5 - 4 -0.26041667 -0.5608547 0.100626889 -2.9973930 0.14893258
6 - 4 -0.20833333 -0.5594223 0.206138515 -2.0776563 0.37679610
6 - 5 0.05208333 -0.3804685 0.465937204 0.4758687 0.99043710
$Overall
Quantile p.Value
1 4.132777 0.02026988
$input
$input$formula
positif.prop.total ~ dose.log
$input$data
dose positif negatif dead totalNb positif.prop.total dose.log
1 0e+00 17 20 0 37 0.45945946 -4.000000e+00
2 0e+00 23 16 0 39 0.58974359 -4.000000e+00
3 0e+00 18 15 0 33 0.54545455 -4.000000e+00
4 0e+00 14 14 1 28 0.50000000 -4.000000e+00
5 1e-03 19 19 1 38 0.50000000 -2.958607e+00
6 1e-03 20 14 4 34 0.58823529 -2.958607e+00
7 1e-02 22 16 0 38 0.57894737 -1.995679e+00
8 1e-02 18 19 0 37 0.48648649 -1.995679e+00
9 1e-02 15 22 2 37 0.40540541 -1.995679e+00
10 1e-02 11 20 4 31 0.35483871 -1.995679e+00
11 1e-01 12 20 0 32 0.37500000 -9.995659e-01
12 1e-01 12 17 4 29 0.41379310 -9.995659e-01
13 1e-01 8 26 3 34 0.23529412 -9.995659e-01
14 1e-01 5 18 11 23 0.21739130 -9.995659e-01
15 1e+00 3 16 10 19 0.15789474 4.342728e-05
16 1e+00 1 16 5 17 0.05882353 4.342728e-05
17 1e+00 2 24 9 26 0.07692308 4.342728e-05
18 1e+00 7 23 6 30 0.23333333 4.342728e-05
19 1e+01 3 10 8 13 0.23076923 1.000004e+00
20 1e+01 2 20 8 22 0.09090909 1.000004e+00
$input$type
[1] "Tukey"
$input$conf.level
[1] 0.95
$input$alternative
[1] "two.sided"
$input$asy.method
[1] "fisher"
$input$plot.simci
[1] FALSE
$input$control
NULL
$input$info
[1] FALSE
$input$rounds
[1] 3
$input$contrast.matrix
NULL
$input$correlation
[1] FALSE
$input$effect
[1] "unweighted"
$input$const
[1] 0.5875441
$text.Output
[1] "True differences of relative effects are not equal to 0"
$text.output.W
[1] "Global Pseudo Ranks"
$connames
[1] "2 - 1" "3 - 1" "4 - 1" "5 - 1" "6 - 1" "3 - 2" "4 - 2" "5 - 2" "6 - 2"
[10] "4 - 3" "5 - 3" "6 - 3" "5 - 4" "6 - 4" "6 - 5"
$AsyMethod
[1] "Fisher with 5 DF"
attr(,"class")
[1] "mctp"
您可以在包 rcompanion 中使用 cldList。您没有提供可重现的数据，所以我将使用 R 中包含的鸢尾花数据集：
data(iris)
library(rcompanion)
library(nparcomp)
iris.mc <- mctp(Sepal.Length~Species, iris)
comp <- iris.mc$connames
pv <- iris.mc$Analysis$p.Value
cldList(comparison=comp, p.value=pv)
# Group Letter MonoLetter
# 1 2 a a
# 2 3 b b
# 3 1 c c