#Welcome to R. A line that starts with the pound sign is a comment
#line. You can type your commands in R line by line or create scripts
#if you prefer. We are going to explore mainly the use of line-by-line
#commands, which is interactive and friendly.

#To get into R from a linux/unix server type R on the command line of a
#shell.  To get to R from a Mac, click on the R icon and then OK on the 
#console. (You have separate instruction files about R logistics on
#the Macs and on Linux\Unix machines).
	
# Assignment. You can create any object in R by using the <- assignment sign.
# Note that _ is equivalent to <- . So, do not use the underscore in the names
# of the objects you create.

x<-1:10

# If you just enter  the name of an R object you see what it consists of.
x

#as you can see : indicates all the numbers between two values.
#There are other ways of generating a sequence of numbers, that one
#can achieve using  the function seq

y<-seq(0.4,35.8,length=10)

#To learn more about seq() (and any other function in R), type

help.start()

# your web-browser will be pointed to a very nice on-line help.
# Note that a function is always invoked with the parenthesis. So, if
# you want to quit R, you need to type

q()

# At this point you will be asked if you want to save the
# workspace. If you say yes, all the objects you have created
# will be saved and re-loaded next time you open R from the same
# directory. Otherwise they will be deleted. It is usually more
# convenient for you to save in a file a copy of the commands to
# create objects, rather than saving objects, which are quite
# space-consuming and not very portable.

# To continue, you will need to start R again.	
	

# Data can be stored into vectors, or matrices, or data.frames.

x<-c(1,2,7,9,20) #is a vector
x[4] # is its fourth element


y<-matrix(c(1,2,3,1,2,3),3,2) # is a matrix with 3 rows and 1
                              # column. Note that it is filled in
                              # column by column.

y[1,]  # is the first row
y[,2]  # is the second column
y[3,1] # is the element on row 3 and column 1

z<-1:5 # is an other vector. You can put it together with x to make a
       # longer vector
c(x,z) # or to make

cbind(x,z) # a matrix

# you can also generate random data

x<-rnorm(100,5,3) #these are 100 random numbers from a normal
                  #distribution with mean 5 and standard deviation
                  #3. You can look for other distributions using the
                  #help for rnorm().

# and, of course, you can load data from a file. Suppose your file
# contains three columns  as the following

# 10 5 A
# 20 6 A
# 1  7 A
# 53 8 A
# 60 1 B
# 2  2 B
# 10 3 B
# 22 4 B

# and suppose that such file is called test. You can load that file in
# R using the function read.table.

x<-read.table(file="test")

# each column will be interpreted as a variable. The last column will
# be interpreted as a categorical variable, a ``factor''.

# One nice thing you can do with your data is to plot it.

plot(x[,1],x[,2]) #scatter plot
hist(x[,1]) #histogram
plot(x[,3],x[,1])  #when the x-coordinate is a factor, you get these
                   #nice box-plots.

# if you want to print your graphs, you need to do a bit of work.
# You start opening the postscript file that will contain them

postscript(file="mygraph",horizontal=F, height=9,width=6) #height and
                                                          #width are
                                                          #in inches
                                                          #and you can
                                                          #change them
#then you execute your plot; this time we want to put them all on the
#same page, so we add an initial command (try with out opening a
#postscript file first, so that you get an idea of what is happening)

par(mfrow=c(3,1))

plot(x[,1],x[,2]) 
hist(x[,1])
plot(x[,3],x[,1])

# and finally you close the file on which you printed.

dev.off()

# Calculating statistics in R is easy:

mean(x[,1]) #gives you the mean of the first column of x.

x<-x[,-3] # we now eliminate the last column of x, so that it is a
          # dataframe of numeric components  
 
apply(x,2,mean) # gives us the column means
apply(x,1,mean) # gives us the row means. Apply is an important
                # function in R. Look it up in the help.

# There are many functions for statistical analysis which are built
# into R and you should always look in the help first, to see if you
# can find something that does what you would like. But sometimes you
# will need to write your own function. This is also rather easy:

myfun<-function(arg1,arg2)
  {
    y<-arg1*3+arg2/3
    return(y)
  }

myfun(1,9)

# R is rather slow in loops, so you should try to avoid loops as much
#as possible. If they cannot be avoided,

test<-NULL
for(i in 1:100)
  {
    test<-c(test,i/6)
    print(i)
  }
test

#of course, you could have avoided this loop by saying

test<-1:100/6

#The following is a web tutorial for Splus. R is not Splus, but it is
#very similar, especially with regard to commands that you can use;
#so you may find useful to consult the following (and I recommend you do)

#http://www.stat.sinica.edu.tw/splus_tutorial/contents.html