If you do not see the menu on the left click here to see it

#Working directory

setwd("H:/public_html/R")

#Identifying the working directory

getwd()

#Memory limit

memory.limit()

#Library --foreign-- to read Stata files

library(foreign)

#Getting data from web

mydata <- read.dta("http://dss.princeton.edu/training/students.dta")

#Extracting variables of interest (optional), here for illustration purposes

library(foreign)

mydata <- read.dta("http://dss.princeton.edu/training/students.dta")

mydata1 <- mydata[c("major","sat")]

mydata1

#Sorting the data by group and value

library(foreign)

mydata <- read.dta("http://dss.princeton.edu/training/students.dta")

#Sorting the data by major

attach(mydata)

mydata <- mydata[order(major, sat),]

detach(mydata)

mydata

mydata1 <- mydata[c("major","sat")]

mydata1

attach(mydata1)

mydata1 <- mydata1[order(major,sat),]

detach(mydata1)

mydata1

##### Getting descriptive statistics per group

library(foreign)

mydata <- read.dta("http://dss.princeton.edu/training/students.dta")

mydata1 <- mydata[c("major","sat")]

mydata1

###Getting the max value by group, in this case by 'major'

max.sat <- with(mydata1, tapply(sat, major, max))

#Adding max variable to the dataset

mydata1$max.sat <- max.sat[as.character(mydata1$major)]

mydata1

#Extracting only data that matches the max value

mydata1.final <- subset(mydata1, sat==max.sat)

mydata1.final

###Getting the mean value by group

mean.sat <- with(mydata1, tapply(sat, major, mean))

#Adding mean.sat variable to the dataset

mydata1$mean.sat <- mean.sat[as.character(mydata1$major)]

mydata1

#Getting the sum value by group, in this case by 'major'

sum.sat <- with(mydata, tapply(sat, major, sum))

#Adding sums per group to the dataset

mydata$sum.sat <- sum.sat[as.character(mydata$major)]

mydata$sum.sat.mis <- sum.sat.mis[as.character(mydata$major)]

#####Creating variables

#Creating variables conditional

mydata1$mean1 <- ifelse(mydata1$sat>mydata1$mean.sat, mydata1$sat, mydata1$mean.sat)

mydata1$mean2 <- with(mydata1, ifelse(sat>mean.sat, sat, mean.sat))

#####Replacing values within variables

#Replacing specific values using conditional

mydata$sat.miss <- ifelse(mydata$sat>1600, mydata$sat, NA)

#Replacing specific value directly

mydata$sat.miss2 <- mydata$sat

mydata$sat.miss2[mydata$sat.miss2<1601] <- NA

#Browsing selected variables

mydata[, c("major","sat", "sat.miss", "sat.miss2", "sum.sat", "sum.sat.mis")]

#Collapse per category

library(foreign)

mydata <- read.dta("http://dss.princeton.edu/training/students.dta")

#Mean by gender (mean on all numeric variables

mean.gender <- aggregate(mydata,by=list(sex=mydata$gender), mean,na.rm=TRUE)

#Sum by major (no collapse)

aggregate(mydata[c("sat")], mydata["major"], sum, na.rm=TRUE)

#Misc

sum.sat.mis <- with(mydata, tapply(sat.miss, major, sum))

#mydata$sat.miss[is.na(mydata$sat.miss)] <- mean(mydata$sat.miss, na.rm = TRUE)

mydata$sat.miss[is.na(mydata$sat.miss)] <- 0

mydata$sat.miss2[is.na(mydata$sat.miss2)] <- 0

sum.sat.mis <- with(mydata, tapply(sat.miss, major, sum))

sum.sat.mis2 <- with(mydata, tapply(sat.miss2, major, sum, na.rm=T))

mydata <- as.data.frame(mydata)

sum <- tapply(mydata$sat,mydata$gender, sum)