Lecture 7 R code - mlloyd23/bio211_JAN2018 GitHub Wiki

##SCRIPT FOR LECTURE 7
##Wrap up transformations

#We will need the rcompanion package, which may require devtools to install
install.packages("devtools")
library(devtools)
install.packages("rcompanion")
library(rcompanion)

###########################################################
##Using river turbidity data example from http://rcompanion.org/handbook/
##########################################################
Input =("
        location turbidity
        a        1.0
        a        1.2
        a        1.1
        a        1.1
        a        2.4
        a        2.2
        a        2.6
        a        4.1
        a        5.0
        a       10.0
        b        4.0
        b        4.1
        b        4.2
        b        4.1
        b        5.1
        b        4.5
        b        5.0
        b       15.2
        b       10.0
        b       20.0
        c        1.1
        c        1.1
        c        1.2
        c        1.6
        c        2.2
        c        3.0
        c        4.0
        c       10.5
        ")

data = read.table(textConnection(Input),header=TRUE)

plotNormalHistogram()

#attempt anova on untransformed data
anova<-


##sqrt transformation
sqrt<-
plotNormalHistogram()

#log
log<-
plotNormalHistogram()

#power transformation performs iterative power transformations
#and re-tests to shapiro.wilk test to optimize the distribution
T_tuk = transformTukey(data$turbidity, plotit = FALSE)
plotNormalHistogram(T_tuk)

data<-cbind()
head(data)

#anova on transformed data
anova.transformed<-

####################################################
##ONTO NON-PARAMETRIC TESTS
####################################################

##Man whitney-u wilxoc ranked sum
##independent t-test on non-normal data
#mtcars dataset 
#disp=displacement which means nothing to me
#because I know nothing about cars
#am is auto (=0) or manual (=1) transmisison

head(mtcars)
hist(mtcars$disp)
shapiro.test(mtcars$disp)
qqnorm(mtcars$disp)
qqline(mtcars$disp)

boxplot()

wilcox.test() 

###The above code is the same as below. Two ways to run the same test
auto<-subset(mtcars, am=="0")

man<-subset(mtcars, am=="1")

wilcox.test(auto$disp,man$disp)

#how does this compare to a traditional t-test?
t.test()

#wilcox signed rank
#paired non-parametric t-test
#back to mpg data
head(mpg)
hist(mpg$cty)
hist(mpg$hwy)

wilcox.test() 

#How does this compared to a traditional t-test?
t.test()

#Kruskal-Wallis 
#ANOVA on non-normal data
head(mpg)
kruskal.test()