R- multiple returns in function to create a plot

Question

I would like to write a function that creates a plot of a set of numbers and adds a regression line. As of now, I can still create the plot but I either get an error with the plot, or some sort of Null response.

My function without a regression line that returns no null looks like this:

fun<-function(){
+   x<-c(1,2,3,4,5)
+   y<-c(1,2,3,4,5)
+   LR<-lm(x~y)
+   
+   return(plot(x,y))
+ }
> fun()

Nice and Pretty with just a plot as a result

if I add the regression line to the plot, I still get the plot but I also get a null response

> fun<-function(){
+   x<-c(1,2,3,4,5)
+   y<-c(1,2,3,4,5)
+   LR<-lm(x~y)
+   p<-plot(x,y)
+   a<-abline(LR)
+   return(p)
+ }
> fun()
NULL

or an error

> fun<-function(){
+   x<-c(1,2,3,4,5)
+   y<-c(1,2,3,4,5)
+   
+   LR<-lm(x~y)
+   
+   p<-plot(x,y)
+   a<-abline(LR)
+   
+   return(p,a)
+ }
> fun()
Error in return(p, a) : multi-argument returns are not permitted

or two nulls

> fun<-function(){
+   x<-c(1,2,3,4,5)
+   y<-c(1,2,3,4,5)
+   
+   LR<-lm(x~y)
+   
+   p<-plot(x,y)
+   a<-abline(LR)
+   
+   return(list(p,a))
+ }
> fun()
[[1]]
NULL

[[2]]
NULL

Sorry if this seems ridiculously nit picky, but in the actual data set it can get ugly.

Justin · Accepted Answer · 2012-06-15 22:29:16Z

5

Does this do what you want?

fun <- function(x, y) {
  plot(x, y)
  abline(lm(x~y))
}

fun(1:10, 10:1)

If you're hoping for a plot object to return for further manipulation, you can use ggplot:

fun <- function(df) {
  ggplot(df, aes(X, y)) + geom_point()
}

df <- data.frame(x=1:10, y=10:1)
p <- fun(df)

# examine p
str(p)

# plot p
p

answered Jun 15, 2012 at 22:29

Justin

43.4k9 gold badges96 silver badges111 bronze badges

Sign up to request clarification or add additional context in comments.

2 Comments

baptiste Over a year ago

+1 for ggplot (I wouldn't recommend str(p) on a complex graphic though ;) ; lattice graphics can also be stored this way.

Justin Over a year ago

@baptiste That's a good point! str(p) will be a big nasty yuk that only Hadley can comprehend. The str(p) was more to make the point that we got something back vs. calling str on the return value of the OP's function.

Thierry · Accepted Answer · 2012-06-15 22:30:57Z

4

Do you really need the function to return the plot object? Or is it sufficient to plot it directly?

I'm addicted to ggplot2 for that kind of stuff. This function does both (printing the plot and returning it as an object)

fun <- function(dataset){
  require(ggplot2)
  p <- ggplot(dataset, aes(x = x, y = y)) + geom_smooth(method = "lm") + geom_point()
  print(p)
  return(p)
}

p1 <- fun(data.frame(x = c(1,2,3,4,5), y = rnorm(5)))
p1

answered Jun 15, 2012 at 22:30

Thierry

18.5k5 gold badges51 silver badges67 bronze badges

Collectives™ on Stack Overflow

R- multiple returns in function to create a plot

2 Answers 2

2 Comments

Comments

Your Answer

Hot Network Questions

Collectives™ on Stack Overflow

2 Answers 2

2 Comments

Comments

Your Answer

Sign up or log in

Post as a guest

Related