##------------------------------------------------------------------------##
## Script for Soc 740  Lecture 1                                          ##
##    John Fox                                                            ##
##    Winter 2014                                                         ##
##------------------------------------------------------------------------##

    # R Basics

# arithmetic, interacting with the interpreter

2 + 3
2 - 3
2*3
2/3
2^3
4^2 - 3*2
(4^2) - (3*2)

1 - 6 + 4

2^-3
-2--3
-2 - -3

# functions, obtaining help

sqrt(4)  # square-root
log(100)  # logarithm
log(100, base=10)
log(100, b=10) # abbreviation
args(log) # function argument
help(log)  # documentation
?log # same as above
example(log) # examples in help page
help.search("log likelihood") # search all installed packages
??"log likelihood" # same as above
RSiteSearch("log likelihood")

log(100, 10) # by position

`+`(2,3) # even operators are functions

# vectors

c(1, 2, 3, 4) # combine

1:4 # sequence operator
4:1
-1:2
seq(1,4) # sequence function
seq(2, 8, by=2)
seq(0, 1, by=.1)
seq(0, 1, length=11)

c(1, 2, 3, 4)/2 # vectorized arithmetic
c(1, 2, 3, 4)/c(4, 3, 2, 1)
log(c(0.1,1,10,100), 10)

c(1, 2, 3, 4) + c(4, 3) # recycling
c(1, 2, 3, 4) + c(4, 3, 2) # warning!

# variables

x <- 1:4 # assignment
x
x = 1:4 # equivalent, to be avoided
x
x/2
y <- sqrt(x)
y
x <- rnorm(100) # destroys original x
x
summary(x) # a 'generic' function

x[21] # indexing
x[11:20]
x[-(11:100)] # drop elements


# user-defined functions

mean(x) 
sum(x)/length(x)

myMean <- function(x) sum(x)/length(x) # function definition
myMean(x)
myMean(y)
myMean(1:100)
x # global variable x untouched

# cleaning up

objects()

remove(x, y, myMean)
objects()

    # Duncan example
    
# creating a data frame from data stored in a file

Duncan <- read.table('D:/data/Duncan.txt', header=TRUE) # assumes location of data
Duncan

Duncan <- read.table(file.choose(), header=TRUE) # a variation
summary(Duncan) # recall 'generic' function

# attaching a data frame
prestige # error!
attach(Duncan) # make data available (there are better ways!)
prestige


# distributions and bivariate relationships

hist(prestige) # histogram

plot(income, education) # scatterplot
identify(income, education, row.names(Duncan)) # point identification
    # to exit point identification: Windows - right-click; Mac or RStudio - Esc
row.names(Duncan)[c(6, 16, 27)]

# fitting a regression

duncan.model <- lm(prestige ~ income + education)
duncan.model # print model object
summary(duncan.model) # again 'generic' function

# using a data frame in an attached package

search()
detach(Duncan)
search()

library(car)
Prestige
View(Prestige) # data viewer

scatterplot(prestige ~ income, data=Prestige) # attach not needed

?scatterplot

scatterplot(income, prestige, span=0.7, labels=row.names(Prestige)) # error!

with(Prestige, # avoiding attach
    scatterplot(income, prestige, span=0.7, labels=row.names(Prestige))) 

scatterplot(prestige ~ income, span=0.7, data=Prestige) # even better (try varying the span)

library(Rcmdr) # graphical user interface, graphs don't play well with RStudio