#list the libraries
library()

#read in built-in dataset
data(stackloss)

#size of data
dim(stackloss)

#variables
names(stackloss)

# first 10 observations
stackloss[1:10,]

# first column
stackloss$Air.Flow
stackloss[,1]

#single entry
stackloss[2,3]
stackloss$s[3]

# R help
help.start()
help(stackloss)
help(dim)
help.search("dimension")

#numerical summary
summary(stackloss)

#mean median quantile
c1=stackloss$Air.Flow
mean(c1)
median(c1)
range(c1)
quantile(c1)
quantile(c1,0.005)
var(c1)
sd(c1)

var
vi(var)
myvar=vi(var)

scale(stackloss$Ai)
sort(c1)
stackloss[order(stackloss$s),]

#correlation + cov
cor(stackloss)
cov(stackloss)

#plots
c1=stackloss$Air.Flow

#histogram
hist(c1)
hist(c1,main="title",xlab="x axis label",ylab="y axis label")
dev.off()
hist(c1,br = 12, col="lightblue", border="pink")
dev.off()

#boxplot
boxplot(c1)
dev.off()
boxplot(c1,notch=T,horizontal=T,col = "bisque")
title("title")
dev.off()

#scatter plot
plot(stackloss$Ai,stackloss$W)
plot(stackloss$Ai,stackloss$W,pch=".",col=5)
plot(Water.Temp ~ Air.Flow, stackloss, xlab="Air Flow",ylab="Water 
Temperature",type="p")
plot(stackloss)
pairs(stackloss)
dev.off()

plot(density(stackloss$Ai))
plot(sort(stackloss$Ai),pch=".")
abline(v=5)
abline(10,7)
points(20,60,col="red",pch=24)

#multiple plots on same page
par(mfrow=c(2,2))
boxplot(stackloss$Ai)
boxplot(stackloss$W)
boxplot(stackloss$Ac)
boxplot(stackloss$s)
par(mfrow=c(1,1))

matplot(stackloss,col=2:5,pch=c(".",":","+","*"))

#vector
v1=c(1,2,4)
v2=c("1","2","4")
v3=5:10
rep(1,5)

#select columns and rows
stackloss[c(1,2,4),]
stackloss[,-1]
stackloss[-c(1:20),]
stackloss[stackloss$Ai>72,]

#----- MATRIX OPERATION---------#

library(MASS)
library(stats)

#read in a file
gala=read.table("../DATA/gala.data",h=T)
mx = gala$Species

#transpose
t(mx)

#matrix multiplication
t(mx) %*% mx
k = mx %*% t(mx)

#inverse
solve(t(mx) %*% mx)

#data type
mm = gala[,-1]
solve(t(mm) %*% mm)
mm = as.matrix(mm)
solve(t(mm) %*% mm)
e=eigen(t(mm) %*% mm) 
e$val
e$vec

diag(mx)
solve(diag(mx))

sqrt(mx)
sum(mx)

#------For Loop------#
test=numeric(10)
for(i in 1:10) {
test[i]=i+rnorm(1)
test[i]=test[i]/(pi^2)
}

abs(test)

#Normal qq plot
t2=rnorm(500)
qqnorm(t2,pch=".",col="blue")
qqline(t2)