[R] Interesting function in a function problem....

[chipmaney]

#I have these data. Basically, I want to run a tapply to calculate the mean
and st. err. by factor.
#The problem is, I want to add a finite correction to the variance prior to
calculating the standard error.
#Now I know I could just do this in 3/4 steps by calculating the var,
applying the correction, then calculating the StErr.

#However, I am trying to learn about functions and how they work. So if you
interested in puzzling the #following problem out..........

#I want to integrate the fc into the tapply() 
#Since the fc is a different object than the tapply() object, how do I
integrate the fc into the function() 
#call for the St.Err so that i get the formula (  fc*var(x)/length(x)  )^0.5
for each factor????
#I believe this takes some fidgeting with function(x), but I suck at
function building, 
#particularly in this case of embedding a function() with multiple inputs
into an R function with only 1 #input,
#so I need help...maybe i need to input the tapply() function into my own
built function???? I don't know.



data <- data.frame(group = c(rep("A",4), rep("B",5)), value =
c(40,45,24,34,234,243,435,324,244) )   
fc <- data.frame(group=c("A","B"), fc = c(0.75, 0.85)) 							

Mean <- tapply(data$value,data$group, mean)
StErr <- tapply(data$value,data$group, function(x) ( var(x)/length(x) )^0.5
) #no fc yet 

-- 
View this message in context: http://www.nabble.com/Interesting-function-in-a-function-problem....-tp25672027p25672027.html
Sent from the R help mailing list archive at Nabble.com.