########################
# Session 2 R code
########################

########################
# R Packages
########################
# To list installed R packages on your computer
library()
# To install a new R package
install.packages("randomForest")
# To load an R package
library("randomForest")
# To list functions and datasets in a package
help(package="randomForest")
# To list data sets in a package
data(package="randomForest")
help(AirPassengers)

#######################
# Read in and Write to External Files
######################
setwd("C:/Users/dzhao1/Desktop")
# Read a text file
RMR = read.table ("RMR.txt", header=T)
# To view the data frame
edit (RMR)
# Read a .csv file
RMR2 = read.csv ("RMR.csv", header=T)

# Write to a .csv file
write.csv (RMR2, "RMRout.csv", row.names=F)



#######################
# Data Management
######################
#Reference: http://www.ats.ucla.edu/stat/r/modules/subsetting.htm
# basic operations on data frames
dim(RMR) #dimensions
names(RMR) #variable names
head(RMR,5) #first 5 obs
tail(RMR,4) #last 4 obs
# or use edit function to view the data frame
edit(RMR)
##### 
# Subset variables rmr and age
#####
# by using indices
RMR.1 = RMR[, c(2,4)]
edit (RMR.1)
# by using variable names
RMR.2 = subset (RMR, select=c(rmr, age))
edit (RMR.2)
##### 
# Subset observations (rows)
# eg, subset all athletes
#####
# by using indices
RMR.3 = RMR[1:8,]
edit(RMR.3)
# or 
RMR.4 = RMR[RMR$athlete==1,]
edit(RMR.4)
# or more naturally
RMR.5 = subset (RMR, athlete==1)
RMR.5
##### 
# Subset observations (rows)
# eg, subset all athletes and age < 30
#####
RMR.6 = subset (RMR, athlete==1 & age <30)
RMR.6

##### 
# Subset observations (rows) and variables (columns)
# eg, subset all athletes and age < 30, keep variable rmr, athlete, age
#####
RMR.7 = subset (RMR, athlete==1 & age <30, select=c(rmr, athlete, age))
RMR.7

##### 
# Merge data frames
#####
# add a column of ID to RMR
RMR=data.frame(ID=1:44, RMR)
RMR.8=subset(RMR,select=c(ID, athlete))
RMR.9=subset(RMR,select=c(ID, age))
RMR.10=merge(RMR.8, RMR.9, by="ID")

########################
# Conditional Excecution
########################

x=2

if (x>0)
{u=1; v=x
} else
{u=0; v=-x}
u
v

#ifelse
u = ifelse(x > 0, sqrt(x), NA)

########################
# Loops
########################
x=4:6
y=x^2
y
rm(y)
# for loops
y=rep(NA,3)
for (i in 1:3) y[i]=x[i]^2
y
rm(y)
# repeat
y=rep(NA,3)
i=1
repeat
{ y[i] = x[i]^2
  if (i==3) break else i=i+1
}
y
rm(y)
# while
y=rep(NA,3)
i=1
while(i<=3){
  y[i]=x[i]^2
  i=i+1
}
y


########################
# R Functions
########################

# to define the function
two.sample.t = function(y1, y2, alpha=.05) {
 n1 = length(y1); n2 = length(y2)
 yb1 = mean(y1); yb2 = mean(y2)
 s1 = var(y1); s2 = var(y2)
 df = n1+n2-2
 s = ((n1-1)*s1 + (n2-1)*s2)/df
 T = (yb1 - yb2)/sqrt(s*(1/n1 + 1/n2))
 p=2*pt(abs(T),df,lower.tail=F)
 LCI=(yb1-yb2)-qt(1-alpha/2,df)*sqrt(s*(1/n1+1/n2))
 UCI=(yb1-yb2)+qt(1-alpha/2,df)*sqrt(s*(1/n1+1/n2))
 CI=c(LCI,UCI)
 list (T=T, p=p, CI=CI, df=df, yb1=yb1, yb2=yb2, y1=y1, y2=y2)
}
# to apply the function
y1=RMR$rmr[1:8]
y2=RMR$rmr[9:44]
z=two.sample.t(y1,y2)

#R built-in function
t.test(y1,y2, var.equal=T)