[R] Plotting linear fit

[Rui Barradas]

Hello,

Try


test <- read.table(text="
  id year incidents
100    1         0
101    1         1
102    1        21
103    1        27
104    1         3
105    1        12
100    2         5
101    2         5
102    2        19
103    2        10
104    2         2
105    2        12
100    3         0
101    3         0
102    3        22
103    3        14
104    3        16
105    3        13
", header=TRUE)


fit <- by(test, test$id, function(dat) fitted.values(lm(incidents ~ 
year, data=dat)))
fit <- unclass(fit)
with(test, interaction.plot(year, id, unlist(fit), col=seq_along(fit)))


Also, it's better to post datasets using dput(): just copy its output 
and paste it.

Hope this helps,

Rui Barradas

Em 22-06-2012 13:20, Van Patten, Isaac T escreveu:
> I am trying to plot the linear fit by id of the following data (test.l):
>
>   id year incidents
> 100    1         0
> 101    1         1
> 102    1        21
> 103    1        27
> 104    1         3
> 105    1        12
> 100    2         5
> 101    2         5
> 102    2        19
> 103    2        10
> 104    2         2
> 105    2        12
> 100    3         0
> 101    3         0
> 102    3        22
> 103    3        14
> 104    3        16
> 105    3        13
>
> I am using this code:
>
> fit<- by(test.l, test.l$id, function(data) fitted.values(lm(incidents ~ year, data=data)))
> fit1<- unlist(fit)
> names(fit1)<- NULL
> interaction.plot(test.l$year, test.l$id, fit1, xlab="Year", ylab="Incidents", ylim=c(0, 40))
>
> Instead of a continuous linear fit, I get a linear fit from year to year.  I am looking to plot
> fit over the entire span of time.  What am I doing wrong?
>
> -Isaac
>
> ______________________________________________
> R-help at r-project.org mailing list
> https://stat.ethz.ch/mailman/listinfo/r-help
> PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
> and provide commented, minimal, self-contained, reproducible code.