## Read comma-separated-version file (csv)
## The first line is the header (header = TRUE)
## The first col contains the row names (row.names = 1)
getwd()
setwd("/u/kalisch/teaching/12/ams/R/v1.1/")
x <- read.csv(file = "USairpollution.csv", header = TRUE, row.names = 1)
## ?read.csv if you need more info on the function "read.csv"

## This help page explains the xa set
?USairpollution

## get an overview of the structure of the data
head(x)
dim(x)
str(x)
##################################################
## mean
##################################################
cm <- colMeans(x)
cm
##################################################
## covariance and correlation
##################################################
## Is there a (linear) association between SO2 and manu?
plot(x$manu, x$SO2) ## Scatterplot
cov(x$manu, x$SO2) ## Covariance: depends on units of measurement
cor(x$manu, x$SO2) ## Correlation: does not depend on that
cor.test(x$manu, x$SO2, method = "pearson") 
## cor.test(x$manu, x$SO2, method = "spearman")

## What about all pairs of variables
plot(x) ## Scatterplotmatrix
S <- cov(x) ## Covariance matrix; save for later use
S
cor(x) ## Correlation matrix

## Parallel Coordinate Plot
parcoord(x)

## Stars Plot
stars(x)
stars(x, key.loc = c(15,1.5), flip.labels = FALSE, draw.segment = TRUE)

## Bubble Plot
plot(x$temp, x$SO2, pch = 20)
symbols(x$temp, x$SO2, circles = x$manu, add = TRUE)

##################################################
## Mixed data
##################################################
library(MVA)
tmp <- read.csv("/sfs/u/staff/kalisch/teaching/12/ams/R/v1.2/australian-crabs.csv")
head(tmp)
dim(tmp)
x <- tmp[,-3]
head(x)
str(x)

specNumeric <- as.numeric(x$species)
sexNumeric <- as.numeric(x$sex)

parcoord(x[,3:7], col = specNumeric)
parcoord(x[,3:7], col = sexNumeric)
## get's easily too crowded ...