#####################################################################
#####################################################################
###
###      Practice num. 1: Starting with R
###                  August 2009                           
###

### R website: http://www.r-project.org/
## R programme, optional packages and manuals can be downloaded from R website.

### Instructions: In this file, every line that starts with "#" is a 
## comment. The rest of lines are R code to be executed in the R console. 

### OPENING R: Double click in R icon and the R console appears. 
## Copy line-by-line the R codes below into the R console, 
## see the results and extract your own conclusions.
## You can add your own comments or explanations.

### 1. DOING SIMPLE CALCULATIONS AND ASIGNING VALUES TO VARIABLES

2*7
30/5
2*(3+4)
sqrt(4)
exp(1)

n <- 15  # Asignment symbol "<-"
n

5 -> x
x

### 2. CHANGE OF DEFAULT WORKING DIRECTORY

# -> File -> Change directory -> [Type or browse your Directory]
# [Close R] -> Save workspace image? Yes
# [Run R from your Directory]

n
x

### 3. CREATION, PRINT OUT AND DELETION OF OBJECTS IN MEMORY

x <- 1 
X <- 10 
x;X   

x1.char <- "I am nice"
x1.char; x

n <- 10 + 2; n

n <- (10+2)*5
n

m <- n + x; m

N <- 1.23e6
N

ls()  
objects()  

M <- m; M
rm(M); M
rm(list=ls())
objects()

name <- "Isabel"
nam
name

n<-10

ls(pattern = "n", all = TRUE)   
ls.str()                       

### 4. HELP IN R

?ls       
help(ls)  
help("ls")  

?survfit   
#Packages->Load package->survival->OK
?survfit   

# Task: Load the package nlme containing the function lme (we will need it) 

help.start() 
apropos("fit") 

### 5. OBJECT ATTRIBUTES

x<-1
mode(x)
length(x)
str(x)
?str

logic.value <- TRUE
mode(logic.value)
length(logic.value )
str(logic.value )

# Task: Find out the mode and length of variable "name" 

### 6. TREATMENT OF NAs AND INCONSISTENT CALCULATION RESULTS

not.available <- NA ; not.available  

# Task: Find out the mode and length of variable "not.available" 

na.number <- c(1,2,NA,4)  # Vector with elements 1, 2, NA and 4
na.number
mode(na.number)
length(na.number)
str(na.number)
na.fail(na.number)  
na.fail(x)
?na.fail
na.omit(na.number)
na.omit(x)

is.vector(x)          
?is.vector
is.vector(na.number)

a <- 256/0
a
exp(a)
exp(-a)
a - a

### 7. OBJETS MANIPULATION

### 7.1. Vectors: Definition and basic operations

x1 <- c(5, 1, 8 ,3)    
order(x1)            
sort(x1)              

x2 <- c(1,2,2,3,3,4)
order(x2)               

x1+x2
x3 <- c(3,2,1)
x2+x3
2*x3     
x2*x3   # Wach out!!        

### Indexation of elements of vectors 

x1[3]                
x2[2]
x1[-4]
x1[-4]+x2
x1
x1[c(1,2)]
x1[-c(1,4)]
x2
x2[x2==3]
x2[x2<3]
x2[x2==3] <-99; x2
x1[order(x1)]          
sort(x1)
  
### Initialization of vectors
                     
a<-vector(mode="numeric",length=6)
str(a)

a<-vector(mode="character",length=6)
str(a)

b<-NULL
b[1]<-2
b[2]<-3

### 7.2. Matrices and Arrays

a1<-matrix(0,nr=5,nc=4);a1
a1 <-array(9,dim=c(5, 4)); a1
dim(a1)
mode(a1)
length(a1)
attributes(a1)

m3 <- matrix(1:20, nrow=5); m3
m3 <- matrix(1:20, nrow=5,byrow=TRUE); m3

# Task: Create a matrix called m4 with 2 rows and 3 columns, with elements c(2,1,4,5,2,3) filled by rows.

z<-3
a2 <- matrix(c(x, z, name), nr=5,nc=4); a2
a2 <- matrix(c(x, z, name), nr=5,nc=4,byrow=TRUE); a2
mode(a2)

a1[1,]
a1[,2]            
a1[1,2]
a1[1,2]<- NA; a1
a1[,4] <- NA; a1

m5 <- 0:19
m5
dim(m5) <- c(5, 4)
m5
m5[1, ]    
m5[1, , drop=FALSE]    
m5[1, , drop=TRUE]    

# Task: Extract the second column of m5: first as a vector, secondly as a submatrix

### Basic operations with matrices

cbind(m3,m5)           
rbind(m3,m5)

m3
t(m3)
m3*m5
m3m5t<-m3%*%t(m5)            
m3m5t

diag(3)
diag(m3)
v <- c(10,20,30)
diag(v)
diag(57,nr=5,nc=7)

Id3 <- diag(3)
solve(Id3)
det(Id3)
eigen(Id3)

### 7.3. Factors

f1 <- factor(c(1, 2, 3)); f1
?factor
f1[2]
str(f1)
is.vector(f1)

f2 <- factor(c(1, 2, 3), labels = c("A", "B", "C"))
f2

f4<-factor(c(1,1,3),levels=c(1,2,3),labels=c("A", "B", "C"))
f4

# Task: Create a factor with elements 1,2,1,2,1,3,3, levels 1,2,3, and labels "lev1","lev2","lev3".

### Data.frames

d1 <- data.frame(g1 = x2, g2 = z); d1  
?data.frame
d1$g2
d1[,2]
d1$g1[2]       
d1[,1]
d1[1]
d1$g3 <- x3; d1
names(d1)

# Task: Create a new data.frame with first column equal to z a with name c1
# and second column equal to x2 and with name c1.
# Then add a new column equal to x3 and with name c3, using function cbind

### 7.4. Lists

L1 <- list(x1, x2, x3)
?list
L1
L1[1]         
L1[[1]]
L1[[1]][3]

# Task: What type of object is the result of L1[1]? And of L1[[1]]

names(L1)

L2 <- list(A=x1, B=x2, C=x3) 
L2

# Task: Extract the first element of L2 as a vector, or as a list
# Task: Extract the third element of vector A in the list L2 

names(L2)
L2$A  
L2$A[3]         

identical(L2$A, L2[1])
identical(L2$A, L2[[1]])

L3 <- list(ls())
L3

# Task: Create a list called L4 with elements x1,x2,a1,a2,d1,name)

### 7.5. Expressions (mathematical functions): Evaluation and derivatives

e1 <- expression(x3 / (x2 + exp(z)))
e1
eval(e1)
D(e1, "z")
D(e1, "x3")
D(e1, "x2")

### 8. DATA GENERATION 

x <- 1:50          
x
x <- -25:25
x
x <- 0.33:68.15024 
x
y<-rep(0,20)
y

1:10-2            
1:(10-2)           

seq(from=23, to=25, by=0.5)  
seq(23, 25, 0.5)  
?seq
seq(length = 100, from = -3e-2, to = 5)
seq(from = -3, by = 0.025, to = 1.86)  

rep(18, 30)        
?rep
rep("Hi", 12)
rep(m4, 3); m4
rep(name, 10)

sequence(4:5) 
sequence(c(2,5,3))   
?sequence

gl(n=4, k=5)               
gl(4, 5)               
gl(n=2, k=5)               
?gl
gl(4, 5, length = 29)  
sex<-gl(2, 6, label=c("Male", "Female"))

# Task: Create a factor with 3 levels, each repeated 10 times
# Task: Create a factor with 3 levels, number of replicates 1, of length 30

f1; f2; c("Male", "Female")
expand.grid(g1 = f1, g2 = f2, gen = c("Male", "Female"))
?expand.grid

# 9. RANDOM SEQUENCES

n <- 50
nvar<-rnorm(n, mean = 0, sd = 1) 
nvar        
?rnorm
evar<-rexp(n, rate = 1)  
evar                 
?rexp
rgamma(10, shape = 1, scale = 1)    
?rgamma
rpois(10, lambda = 0.5)             
?rpois

### 10. SIMPLE ARITHMETIC AND STATISTICAL FUNCTIONS

x1
x2
y <- x2[-c(5, 6)]
y
sum(x1)
prod(x1)
max(x1)
min(x1)
which.max(x1)
which.min(x1)
range(x1)
length(x1)
mean(x1)
median(x1)
var(x1)
cov(x1, x1)
cor(x1, x1)
var(x1, y)
cov(x1, y)
cor(x1, y)
summary(x1)

# Task: Calculate the mean and variance of the random vector nvar
# Task: Calculate the correlation between random vectors nvar and evar
# Task: Do a summary of the normal vector nvar
# Task: Do a summary of the factor sex and observe the result

### 11. SIMPLE STATISTICAL PLOTTING FUNCTIONS

plot(nvar,evar)

hist(nvar)
hist(nvar,prob=TRUE)
boxplot(nvar)

x<-seq(-3,3,by=0.2)
plot(x,dnorm(x,0,1))
plot(x,dnorm(x,0,1),type="l")
plot(x,dnorm(x,0,1),type="b")

plot(x,dnorm(x,0,1),type="n")
points(x,dnorm(x,0,1),type="l",lwd=2)
points(x,dnorm(x,0,1.5),type="l",col=2,lwd=2)
abline(h=0.1)
abline(v=-2)

# Task: Plot the histogram of evar. 
# Then superpose a line of the density of an Exponential variable 
# with the same parameter value (rate)

### 12. READING AND WRITING DATA

### Reading from keyboard

a <- scan()
1 2 3 4 5
6 7 8 9 10

?scan

### Reading from a file

data<-read.table("silc0106_ISI09.txt",header=TRUE)
data[1:10,]
prov[1:10]
data$prov[1:10]
data[,2][1:10]
attach(data)
prov[1:10]
?read.table

# Task: Read the file PopnSizeProv.txt and see the population sizes of Spanish provinces

### Writing results to a file

x1 <- rnorm(100)
x2 <- rnorm(100)
x3 <- rnorm(100)
x4 <- rnorm(100)
x5 <- rnorm(100)
a <- data.frame(a=x1, b=x2, c=x3, d=x4, e=x5)

write.table(a, file = "a.txt", row.names = FALSE)
?write.table

write.table(a, file = "a.xls", row.names = FALSE, sep="\t")

sink(file = "sinkfile.txt")
     i <- 1:10 ; i
     sim1 <- rnorm(100)
     sim1
     sim2 <- rpois(100,1)
     sim2
sink() 
?sink