Skip to content

boot error: Error in statistic(data, original, ...) : unused argument(s) ( ...)

3 messages · John Sorkin, Brian Ripley, Thomas Lumley

#
I have added on additional line to the code below, the line the
defines the matrix passed to AdjForBase2.

I am having difficulty setting up a boot strap.
My code is listed below. I am getting the following
error message:
Error in statistic(data, original, ...) : unused argument(s) ( ...)
I neither understand the error I have made that leads to 
the error message, nor do I know how to fix my error.
Any help would be appreciated.
Thank you,
John
+ function (data) 
+ {
+ cat("******************\n")
+ #print(length(x))
+ 
+ x1<-data[,1]
+ x2<-x1+rnorm(100,0,2)
+ xdiff<-x2-x1
+ corx1x2minx1<-cor(x1,xdiff)
+ 
+ y1<-data[,2]
+ y2<-y1+rnorm(100,10,2)
+ ydiff<-y2-y1
+ cory1y2miny1<-cor(y1,ydiff)
+ 
+ return(c(mean(x1),mean(x2),corx1x2minx1,
+          mean(y1),mean(y2),cory1y2miny1))
+ }
Error in statistic(data, original, ...) : unused argument(s) ( ...)
#
Please look at the examples and the book which package boot supports (see 
its DESCRIPTION file).  From ?boot

statistic: A function which when applied to data returns a vector
           containing the statistic(s) of interest.  When
           'sim="parametric"', the first argument to 'statistic' must be
           the data.  For each replicate a simulated dataset returned by
           'ran.gen' will be passed.  In all other cases 'statistic'
           must take at least two arguments.
           ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
and yours only takes one argument, hence the error message.

I suspect you want

AdjForBase2 <- function (data, inds)

and to refer to data[inds, 1] and data[inds, 2], but since your code is 
completely devoid of spaces and indentation, I have paid it little 
attention.
On Fri, 12 Aug 2005, John Sorkin wrote:

            

  
    
2 days later
#
On Sat, 13 Aug 2005, Prof Brian Ripley wrote:
http://msr.uwaterloo.ca/msr2005/papers/27.pdf  describes a study of 
student computer science projects from CVS logs in which the only code 
feature correlated with the final grade was the number of times a space 
was used after a comma.  [I'm not sure I like their analysis, though].


 	-thomas