一、R函数

1. 线性回归

state <- as.data.frame(state.x77[,c("Murder" ,"Population","Illiteracy","Income", "Frost")])
fit<-lm(Murder~Population+Illiteracy+Income+Frost,data=state)
summary(fit)

可以看出在P<0.001的情况下，Illiteracy的系数显著不为0
其他因子没有显著不为0，判断无相关关系

2. 概率分布函数

##正态分布 
?Normal
rnorm(50, mean = 0, sd = 1)
x<-rnorm(50, mean = 0, sd = 1)
qqnorm(x)

##Beta分布 
?beta 
##Logistic分布 
?logis 
##二项分布 
?binom.test()
##多项分布 
?multinom 
##柯西分布 
cauchy 
##负二项分布 
?nbinom 
##卡方分布 
?chisq 
##指数分布 
?exp 
##泊松分布 
?pois 
##F分布 
?f 
##符号秩分布 
?signrank 
##Gamma分布 
?gamma  
##几何分布 
?geom 
##均匀分布 
?unif 
##超几何分布 
?hyper 
##Weibull分布 
?weibull 
##对数正态分布 
Inorm 
##Wilcoxon秩和分布 
?wilcox

如何检验数据集满足某种分布
后续更新

3. 生成随机数

?runif
runif(1)
runif(50)
#生成50个，1-100的随机数
runif(50,min=1,max=100)
set.seed(111)
runif(50)
#绑定seed，可以生成相同随机数
set.seed(111)
runif(50)

4. 描述性统计函数

描述分布

summary(myvars)
fivenum(myvars$hp)
tDescriptive stats via stat.desc (pastecs) install.packages("pastecs") library(pastecs)
stat.desc(mtcars[myvars])
stat.desc(mtcars,basic = TRUE,desc=TRUE,norm = TRUE)
library(psych)
describe(mtcars[myvars],trim =0.1) Hmisc::describe(mtcars)

聚合后分布

library(MASS)
aggregate(Cars93[c("Min.Price","Price","Max.Price","MPG.city")],
by=list(Manufacturer=Cars93$Manufacturer),mean)
aggregate(Cars93[c("Min.Price","Price","Max.Price","MDG.city")],
byslist(Manufacturer=Cars93$Manufacturer),sd)
aqareqate(Cars93[c("Min.Price","Price","Max.Price","MPG.city")],
by=list(origin=Cars93$origin),mean)
agqregate(Cars93[c("Min.Price","Price" "Max.Price","MPG.city")],
by=list(origin=Cars93$origin) ,sd)
aggregate(Cars93[c("Min.Price","Price","Max.Price","MPG.city")],
by=list(origin=Cars93$origin,Manufacturer=Cars93$Manufacturer),mean]

根据自变量因子分类后，其他因子分布描述

library(doBy)
summaryBy(mpg+hp+wt~am,data=mtcars,FUN=mean)
myvars <- c("mpg","hp","wt")
library(psych)
describeBy(mtcars[myvars],list(am=mtcars$am))

5. 频数统计函数

#分组
mtcars$cyl<-as.factor(mtcars$cyl)
split(mtcars,mtcars$cyl)
split(mtcars,as.factor(mtcarsScyl))
#切割
num <-1:100
cut (num,c(seq(0,100,10)))
cut (mtcars$mpg,c(seq(10,50,10))) 
##频数统计
table(mtcars$cyl)
table(cut(mtcars$mpg,c(seq(10,50,10))))
##频率函数
prop.table(table(mtcars$cyl))

> library("grid")
> library("vcd")
> table(Arthritis$Treatment,Arthritis$Improved)
> ?margin.table
x <- xtabs(~ Treatment +Improved,data= Arthritis)
margin.table(x)
margin.table(x,1) 
prop.table(x,1)