Skip to content

Fitting polynomial (power greater than 2)

2 messages · dkendall44, Greg Snow

Original
dkendall44
# 
Hey all, first time poster here. I'm new to R and working on my first real
programming and forecasting asignment. I'm using unemployment data from
1948-2012.  I successfully completed part a and the linear fit for part b,
but i am really struggling fitting a polynomial with a power greater than 2
to my forecast. I'll upload my R code at the bottom. Any help is very much
appreciated! Thank You!

(a) Produce a time-series plot of your data.
 (b) Fit linear, polynomial (use a power greater than 2), and R?s best fit
models to your series.

rm(list=ls(all=TRUE))
z=read.table("longtermunempdata.txt")
names(z)= c("Unemployment")
attach(z)
TIME=seq(1,769,by=1)

Unemployment_ts<-ts(Unemployment,start=1948,freq=12)
#quartz()
#plot(stl(Unemployment_ts,s.window="periodic"))

#1A
quartz()
plot(TIME, Unemployment, xlab="Year", ylab="Long Term
Unemployment",type='l')

#1B Linear
quartz()
plot(TIME,Unemployment,type="l", xlab="Year",ylab="Long Term Unemployment")
modelu=lm(Unemployment~TIME)
summary(modelu)
abline(modelu,col="red3")

modelu=lm(Unemployment~TIME)
#1B Linear vs TS
quartz()
par(mfrow=c(2,1))
plot(TIME, Unemployment, xlab="Year", ylab="Long Term
Unemployment",type='l')
abline(modelu,col="red3",lwd=2)
plot(TIME,modelu$res,xlab="Year",ylab="Long Term Unemployment",type='l')

--
View this message in context: http://r.789695.n4.nabble.com/Fitting-polynomial-power-greater-than-2-tp4368026p4368026.html
Sent from the R help mailing list archive at Nabble.com.
1 day later
Greg Snow
# 
This looks like homework, which is generally discouraged here.  You
should at the least admit that it is homework and tell what resources
your teacher has OK'd you to use.

Since you did show what you have done so far and are not just asking
for a handout, here is a hint:

?poly
On Tue, Feb 7, 2012 at 9:34 PM, dkendall44 <dkendall44 at gmail.com> wrote: