Prague Half Marathon 2010

pim.R

library(xlsReadWrite)
library(MASS)

### DATA DOWNLOAD

dir.name <- tempdir()

url2010 <- "http://www.praguemarathon.com/images/stories/articles/files/HM2010/HM2010_eng.xls"
file <- paste(dir.name, "2010.xls", sep="\\")
download.file(url2010, file, mode="wb") 
data2010 <- read.xls(file)
names(data2010)[1] <- "number"
data2010 <- data2010[-2471,] #Realtime = 0:00:00

url2009 <- "http://behycz.s3.amazonaws.com/media/files/vysledky/2113-hervis-1-2maraton-praha.xls?1299838523"
file <- paste(dir.name, "2009.xls", sep="\\")
download.file(url2009, file, mode="wb") 
data2009 <- read.xls(file)
names(data2009) <- names(data2010)

url2011 <- "http://behejcom.s3.amazonaws.com/media/files/other/neoficialni%20vysledky%20Hervis%20pulmaratonu%202011.xls?1301767966"
file <- paste(dir.name, "2011.xls", sep="\\")
download.file(url2011, file, mode="wb") 
data2011 <- read.xls(file)
names(data2011)[1:16] <- names(data2010)[1:16]

### START TIME COMPUTATION

data2010$start.time <- strptime(data2010$Offic..time,"%H:%M:%S") - strptime(data2010$Real.time,"%H:%M:%S")
data2010$start.time <- as.numeric(data2010$start.time)  

data2009$start.time <- strptime(data2009$Offic..time,"%H:%M:%S") - strptime(data2009$Real.time,"%H:%M:%S")
data2009$start.time <- as.numeric(data2009$start.time)  

data2011$start.time <- strptime(data2011$Offic..time,"%H:%M:%S") - strptime(data2011$Real.time,"%H:%M:%S")
data2011$start.time <- as.numeric(data2011$start.time)

### HISTOGRAMS

with(data2009, hist(start.time,breaks=seq(from=-10,to=max(start.time),by=1), main="Histogram of start times in 2009", xlim=c(0,570), ylim=c(0,250), xlab="Start time [s]"))
abline(h=20, col="red", lty=2)
with(data2010, hist(start.time,breaks=seq(from=0,to=max(start.time),by=1), main="Histogram of start times in 2010", xlim=c(0,570), ylim=c(0,250), xlab="Start time [s]"))
abline(h=20, col="red", lty=2)
with(data2011, hist(start.time,breaks=seq(from=0,to=max(start.time),by=1), main="Histogram of start times in 2011", xlim=c(0,570), ylim=c(0,250), xlab="Start time [s]"))
abline(h=20, col="red", lty=2)

### ORDERING TO CORRIDORS

with(subset(data2009, Sex=="M"& as.numeric(number)<3000), plot(start.time ~ as.numeric(number), cex=0.1, main="Ordering to corridors in 2009", ylim=c(0,600), xlab="Start number", ylab="Start time [s]"))
with(subset(data2010, Sex=="M" & as.numeric(number)<3000), plot(start.time ~ as.numeric(number), cex=0.1, main="Ordering to corridors in 2010", ylim=c(0,600), xlab="Start number", ylab="Start time [s]"))
with(subset(data2011, Sex=="M"& as.numeric(number)<3000), plot(start.time ~ as.numeric(number), cex=0.1, main="Ordering to corridors in 2011", ylim=c(0,600), xlab="Start number", ylab="Start time [s]"))
abline(rlm(start.time ~ as.numeric(number), data=subset(data2011, Sex=="M"& as.numeric(number)<3000)), col="blue", lty=4)