简体   繁体   English

R:基本功能无法使用Caper包中的对象

[英]R: Base functions cannot use object from the package Caper

I'm running a phylogenetic analysis using the caper package, where the regression function (which uses phylogeneticaly independent contrasts) is crunch . 我正在使用caper包进行系统发育分析,其中的回归函数(使用系统发育独立的对比)非常crunch The crunch function uses an object internal to the caper package called caic . crunch功能使用在caper包内部的名为caic的对象。

The model is started via: 该模型通过以下方式启动:

crunchMod <- crunch(y ~ f(x), data = comparison)

When I run summary(crunchMod) I am given a format identical to that of a summary produced from the lm() function. 当我运行summary(crunchMod)时,会得到与lm()函数生成的摘要相同的格式。

However, in an attempt to begin checking model assumptions by entering rstandard(crunchMod) , I receive the following error: 但是,尝试通过输入rstandard(crunchMod)开始检查模型假设时,出现以下错误:

Error in UseMethod("rstandard") : 
no applicable method for 'rstandard' applied to an object of class "caic"

Reading through http://cran.r-project.org/web/packages/caper/vignettes/caper.pdf on pages 19-20, I find that plot(crunchMod) uses a wrapper to enable the checking of regression assumptions. 通过阅读第19-20页的http://cran.r-project.org/web/packages/caper/vignettes/caper.pdf ,我发现plot(crunchMod)使用包装器可以检查回归假设。 However, these are graphical checks for: 但是,这些是图形检查:

residuals vs Fitted values
standardized residuals vs theoretical quantities [QQ plot]
sqrt(standardized residuals) vs fitted values [Scale-location]
standardized residuals vs leverage).

Does anyone know how to access the standardized residuals using my own wrapper, or allow me to access p-values instead of the graphical images? 有谁知道如何使用我自己的包装器访问标准化残差,或者允许我访问p值而不是图形图像?

This one is pretty simple, but only after going round in circles for a while with that second if() statement in crunch(). 这是非常简单的,但只有在使用crunch().中的第二条if()语句绕了一圈之后才进行crunch(). Looking at the summary method for caic , it's just a subset of the entire summary / model 查看caic的摘要方法,它只是整个摘要/模型的一部分

> summary.caic
function (object, ...) 
{
    summary(object$mod, ...)
}
<environment: namespace:caper>

You can see the names of the entire model show there are useful statistics in the rest of it. 您可以看到整个模型的名称,其中显示了其余的有用统计信息。

names(summary(crunchMod))
# [1] "call"          "terms"         "residuals"     "coefficients" 
# [5] "aliased"       "sigma"         "df"            "r.squared"    
# [9] "adj.r.squared" "fstatistic"    "cov.unscaled" 

Note that only the mod section inherits from lm() You can check out all the inheritence of the crunchMod objects with 请注意,只有mod部分是从lm()继承的,您可以使用以下命令检查出crunchMod对象的所有继承

> sapply(crunchMod, is)
$contrast.data
[1] "list"   "vector"

$mod
[1] "lm"       "oldClass"

$data
[1] "comparative.data"

声明:本站的技术帖子网页,遵循CC BY-SA 4.0协议,如果您需要转载,请注明本站网址或者原文地址。任何问题请咨询:yoyou2525@163.com.

 
粤ICP备18138465号  © 2020-2024 STACKOOM.COM