Documents/R/testing-baseline.R

In thl-mjv/euromomo: EuroMoMo algorithm

#loadDefaults(TRUE)

#options()$euromomo

data <- read.csv2("delay.txt")
tail(data)
data$ISOweek<-with(data,sprintf("%04d-W%02d",YoDi,WoDi))
data2<-addconditions(data,spring=15:26,autumn=30:48,delay=1:2)
summary(data2)

data3 <- baseline(data2)
#                  ,seasonality=TRUE)

head(data3)
with(data3,as.data.frame(table(CondSeason,CondSomething,CondDelays,CondLength,cond)))


data4 <- zscore(data3)

data5 <- excess(data4,type="baseli")

tail(data5)

library("lattice")
names(data5)
xyplot(nb+cnb+pnb+u.pnb+l.pnb~wk,data=data5,subset=wk>30,
       type="l",ylim=c(0,2000),lty=c(1,1,1,2,2,2,2),lwd=c(2,2,1,1,1,1,1),
       col=c(1,2,3,3,3,2,2))
xyplot(Zscore~wk,data=data5,type="l")
xyplot(excess+u.excess+l.excess~wk,data=data5,type="l")

#### the problem with spines
x1<-1:200
x2<-1:250
library(splines)
matplot(x1,bs(x1),type="l",xlim=c(1,250))
matlines(x2,bs(x2),type="l")