用“R”编程语言显示和报告函数输出

在R编程语言和包pcaPP中，我有以下代码：

#  multivariate  data  with  outliers
library(mvtnorm)
library(pcaPP)

x  <-  rbind(rmvnorm(200,  rep(0,  6),  diag(c(5,  rep(1,5)))),
             rmvnorm(  15,  c(0,  rep(20,  5)),  diag(rep(1,  6))))
#  Here  we  calculate  the  principal  components  with  PCAgrid
pc  <-  PCAproj(x)

---

如何调用PCAproj的其他输出，如loadings和sdev，并在R-studio中报告这些输出？

在您的示例中，它们都存储在pc中

如果您处于交互模式，只需键入pc$sdev和pc$loading即可查看它们包含的内容

> pc$sdev
  Comp.1   Comp.2
2.425413 1.346727


> pc$loadings

Loadings:
   Comp.1 Comp.2
V1  0.972  0.153
V2 -0.201  0.447
V3        -0.130
V4        -0.211
V5         0.739
V6 -0.109  0.412

               Comp.1 Comp.2
SS loadings     1.000  1.000
Proportion Var  0.167  0.167
Cumulative Var  0.167  0.333

---

为了补充一下Bottoms先生所说的内容，我发现以下函数集在深入挖掘pc对象的输出（名称、str和摘要）时非常有用

str是structure的缩写。它输出一个易于阅读的R数据结构描述。我喜欢把它看作一个目录。您会注意到它与您的姓名列表相匹配

str(pc)
List of 7
 $ loadings: loadings [1:6, 1:2] 0.962 0.1011 0.048 0.2461 0.0152 ...
  ..- attr(*, "dimnames")=List of 2
  .. ..$ : chr [1:6] "V1" "V2" "V3" "V4" ...
  .. ..$ : chr [1:2] "Comp.1" "Comp.2"
 $ sdev    : Named num [1:2] 2.79 1.39
  ..- attr(*, "names")= chr [1:2] "Comp.1" "Comp.2"
 $ center  : num [1:6] 0.193 0.114 0.093 0.117 0.215 ...
 $ scale   : num [1:6(1d)] 1 1 1 1 1 1
 $ n.obs   : int 215
 $ scores  : num [1:215, 1:2] -0.413 1.707 0.835 2.164 0.495 ...
  ..- attr(*, "dimnames")=List of 2
  .. ..$ : chr [1:215] "1" "2" "3" "4" ...
  .. ..$ : chr [1:2] "Comp.1" "Comp.2"
 $ call    : language PCAproj(x = x)
 - attr(*, "class")= chr [1:2] "pcaPP" "princomp"

大多数设计良好的函数允许您将新对象传递到summary函数中，它将返回。。。让我们称之为该函数输出的最明显和最有用的摘要

> summary(pc)
Importance of components:
                          Comp.1    Comp.2
Standard deviation     2.7873357 1.3855889
Proportion of Variance 0.8018539 0.1981461
Cumulative Proportion  0.8018539 1.0000000

然后RStudio和其他ide有一些很酷的功能，比如tab auto complete，所以如果你键入pc$，然后点击tab键，它会列出上面列出的所有名称。然后可以使用箭头键选择要选择的元素

> str(pc$loadings)
 loadings [1:6, 1:2] 0.962 0.1011 0.048 0.2461 0.0152 ...
 - attr(*, "dimnames")=List of 2
  ..$ : chr [1:6] "V1" "V2" "V3" "V4" ...
  ..$ : chr [1:2] "Comp.1" "Comp.2"

---

通过执行strpc检查pc的结构，您可能会得到一些想法

str(pc)
List of 7
 $ loadings: loadings [1:6, 1:2] 0.962 0.1011 0.048 0.2461 0.0152 ...
  ..- attr(*, "dimnames")=List of 2
  .. ..$ : chr [1:6] "V1" "V2" "V3" "V4" ...
  .. ..$ : chr [1:2] "Comp.1" "Comp.2"
 $ sdev    : Named num [1:2] 2.79 1.39
  ..- attr(*, "names")= chr [1:2] "Comp.1" "Comp.2"
 $ center  : num [1:6] 0.193 0.114 0.093 0.117 0.215 ...
 $ scale   : num [1:6(1d)] 1 1 1 1 1 1
 $ n.obs   : int 215
 $ scores  : num [1:215, 1:2] -0.413 1.707 0.835 2.164 0.495 ...
  ..- attr(*, "dimnames")=List of 2
  .. ..$ : chr [1:215] "1" "2" "3" "4" ...
  .. ..$ : chr [1:2] "Comp.1" "Comp.2"
 $ call    : language PCAproj(x = x)
 - attr(*, "class")= chr [1:2] "pcaPP" "princomp"

> summary(pc)
Importance of components:
                          Comp.1    Comp.2
Standard deviation     2.7873357 1.3855889
Proportion of Variance 0.8018539 0.1981461
Cumulative Proportion  0.8018539 1.0000000

> str(pc$loadings)
 loadings [1:6, 1:2] 0.962 0.1011 0.048 0.2461 0.0152 ...
 - attr(*, "dimnames")=List of 2
  ..$ : chr [1:6] "V1" "V2" "V3" "V4" ...
  ..$ : chr [1:2] "Comp.1" "Comp.2"