[R] errors appears in my time Series regression fomula

[snowcat wong]

Dear all,

 

I found really difficult with the time series questions, please help me with this monthly airline series!

 

I have run the following r code, and there is an error appeared at the end.  The data files was enclosed in the email.

I'm sorry the errors message appeared in chinese, but it says "plot.xy(xy.coords(x, y), type = type, ...) :

errors in argument has more than 3 matches"

 

> y=scan('F:/Math355/AIRLINE.txt',skip=2,nlines=18)
Read 144 items
> yts=ts(log(y))
> plot(yts,main="LOG AIRLINE TOTALS",type='p',col=2)
> n=length(y)
> time=seq(1:n)
> 
> month=c(rep(seq(1:12),12))
> fmonth=as.factor(month)
> ymod1=lm(yts~time+fmonth)
> summary(ymod1)

 

> yfit1=ymod1$fitted
> lines(yfit1,c=3)
错误于 plot.xy(xy.coords(x, y), type = type, ...) : 
  变元3有多个与之相对应的正式变元


Also, for a similar data set quarterly sales of spirits series, what is the best model to use?? Should I use ARIMA??

 

 

 

Many thanks and regard,

Cathy Wong
 		 	   		  
_________________________________________________________________
Got a cool Hotmail story? Tell us now
-------------- next part --------------
An embedded and charset-unspecified text was scrubbed...
Name: spirits Rdata.txt
URL: <https://stat.ethz.ch/pipermail/r-help/attachments/20100118/c1be065b/attachment.txt>
-------------- next part --------------
An embedded and charset-unspecified text was scrubbed...
Name: SPIRITS.txt
URL: <https://stat.ethz.ch/pipermail/r-help/attachments/20100118/c1be065b/attachment-0001.txt>