R - Median of a Frequency distribution, grouped by another variable

Question

I have a data set that look like the following: http://i.imgur.com/OdiLf4t.png

Name | State | Zipcode | County_name | average payment | Frequency

My desired output would be to group by State and have the Median payment using the average payment and Frequency columns.

I know how to do this for the overall dataset

median(rep(Clean$medicare_average_payment, Clean$Frequency))        

but not sure how to do this by State Thank you

Answer 1

We can try with dplyr

library(dplyr)    
Clean1 <- Clean[rep(1:nrow(Clean), Clean$Frequency),]
Clean1 %>%
      group_by(State) %>%
      summarise(Median = median(medicare_average_payment))

---

Or using data.table

library(data.table)
setDT(Clean)[, .(Median = median(rep(medicare_average_payment, Frequency))) , State]

Answer 2

You can use by to do split the data frame and perform this function on each piece:

by(Clean, Clean$State, 
   FUN=function(x) median(rep(x$medicare_average_payment, x$Frequency))
)