################ Example 1: Data are attributes on mammalian morphology and reproduction #########################

# read the data from a file and into a data frame (enhanced matrix)

Data <- read.table(file="C:/Documents and Settings/steeleb/My Documents/Stat 341/BrainWeight.txt",sep = ",",header=T)
# print the first 6 rows of the data frame
head(Data)
colnames(Data)

print(Data$species)
print(Data[,1])    #print column 1 of the data frame
Data[,1]
Data[1,]		#print row 1 of the data frame
attach(Data)	# make each column into a vector with the same name as the column
species		# print the species names

plot(body,gestation)
plot(log(body),gestation)	# transform body weight to the natural logarith scale before plotting
identify(log(body),gestation,species) # note: hit Esc key to stop labelling points

plot(log(body),gestation,pch=16,col="red")
palette()
plot(body,brain,pch=16,col="green")
plot(log(body),log(brain),pch=16,col="black")

############################# construct new data.frame

new.data <- data.frame(species,body,brain,row.names=1)
head(new.data)
detach(Data) # eliminate the variables from R memory

x <- seq(from=1,to=10,by=1)
y <- rep(1:5,2)

data.frame(x,y)
names <- c("a","b","c","d","e","f","g","h","i","j")
length(names)
new.data <- data.frame(names,x,y,row.names=1)
new.data
colnames(new.data)

######################### use data from a library ##############
library(DAAG)

data <- Manitoba.lakes
head(data)
log(data$area)
row.names(data)
plot(data$elevation,log(data$area))
#################### Example 2: Data are number of TB cases per 100,000 people #########################

# read the data from a file
Data <- read.table(file="C:/Documents and Settings/steeleb/My Documents/Stat 341/GlobalTB.txt",header=T)

# remove rows that contain missing data (represented by NA)
Data <- na.omit(Data)

# extract the names of the columns
column.names <- names(Data)
# print the column names
column.names

# print the first 10 rows of the data frame
head(Data,10)

# print the dimension of the data frame
dim(Data)
# store the number of rows (countries) and columns
n <- dim(Data)[1]
p <- dim(Data)[2]

# store the TB rates leaving off the country and region names
X <- Data[,3:p]

# store the country names
Country <- Data$Country
# store the region names
Region <- Data$Region
# construct a list of the regions
regions <- levels(Region)

# construct a sequence corrresponding to the years of observation
Year <- 1982:2005
Year

# get the data for Algeria (first country in the data.frame
TB.cases.Algeria <- X[1,]
plot(Year,TB.cases.Algeria)
plot(Year,TB.cases.Algeria,type="b",pch=16)

## use a high-level graphics library - tells R to open the package
library(lattice)

plot(c(1982,2005),c(0,max(X)),type="n",ylab="Rate per 100,000",xlab="Year") # create a plotting field
I <- Country=="Zimbabwe"
I
lines(Year,X[I,],col=1) # add a country
I <- Country=="Botswana"
lines(Year,X[I,],col=2)  # add a country
I <- Country=="SouthAfrica"
lines(Year,X[I,],col=3)

I <- Country=="Philippines"
lines(Year,X[I,],col=4)
I <- Country=="India"
lines(Year,X[I,],col=5)
I <- Country=="SKorea"
lines(Year,X[I,],col=6)

legend(x="topleft",legend=c("Zimbabwe","Botswana","SouthAfrica","Philippines",
"India","SKorea"),col=1:6,lty=1)