[英]GLM in R: Normal Q-Q Plot with the 95% confidence interval
I'd like a function or package to plot the Normal QQ Plot with the 95% confidence interval, but I don't find for GLM, only GAM models and for response variables in package car
.我想要一个函数或包来绘制具有 95% 置信区间的正态 QQ 图,但我没有找到 GLM,只有 GAM 模型和包
car
响应变量。 In my example:在我的例子中:
#Data set example
p <- read.csv("https://raw.githubusercontent.com/Leprechault/PEN-533/master/bradysia-greenhouse.csv")
#Quasi Poisson GLM
m1 <- glm(bradysia ~ area + mes, family="quasipoisson", data=p)
#Normal Q-Q Plot
plot(m1, which = 2)
#Normal Q-Q Plot with the confidence interval
I'll like a plot like above for GLM models, it is possible?我会喜欢上面 GLM 模型的图,这可能吗? There are any function or package?
有什么函数或包吗?
There are two types of confidence intervals: 1.confidence interval for prediction (what you have shown in your plot 2) and 2. confidence interval for regression.有两种类型的置信区间:1. 预测的置信区间(您在图 2 中显示的内容)和 2. 回归的置信区间。 To fit confidence interval for regression to the residuals of the glm model you can use
car
package like要拟合回归到 glm 模型残差的置信区间,您可以使用
car
包,如
library(car)
qqPlot(m1$resid, ylab="Residuals", xlab="Theoretical Quantiles")
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.