Aggregate concept in R

Asked Oct 05 '16 at 05:29

Active Oct 05 '16 at 05:37

Viewed 18 times

I have a dataframe mk

    HomeID  tract     value
1        1 FIPS04 314522.70
2        2 FIPS03 201834.99
3        3 FIPS03 196524.75
4        4 FIPS03 198746.18
5        5 FIPS04 296456.83
6        6 FIPS02 130258.99
7        7 FIPS02 122678.42
8        8 FIPS02 152234.58
9        9 FIPS03 164635.87
10      10 FIPS02 131209.42

I have to aggregate the dataframe by tract and put those values into a new dataframe and calculate the aggregated values of mean,sd,length.

edited Oct 05 '16 at 05:37

Ronak Shah

377,200
20
156
213

asked Oct 05 '16 at 05:29

Rohit Venkat Gandhi Mendadhala

Just do `aggregate(value~tra, mk, mean)` For multiple functions `f1 <- function(x) c(mean(x), sd(x),length(x));do.call(data.frame, aggregate(value~tra, mk, f1))` – akrun Oct 05 '16 at 05:30
The factor given was tra... but not value.. – Rohit Venkat Gandhi Mendadhala Oct 05 '16 at 05:31
If you are grouping by 'tra', that is the syntax or use `library(dplyr);mk %>% group_by(tra) %>% summarise_each(funs(mean, sd, n()), value)` According to your post `I have to aggregate the dataframe by tra` – akrun Oct 05 '16 at 05:32
The first parameter.. is the factor by which we should aggregate right? So as per my question it should be value or tra? – Rohit Venkat Gandhi Mendadhala Oct 05 '16 at 05:33
It is grouping by 'tra' and getting the `mean`, `sd` etc of 'value'. Does it makes sense to get the `mean` of 'tra'? – akrun Oct 05 '16 at 05:34
Ok, is it not required to have HomeID – Rohit Venkat Gandhi Mendadhala Oct 05 '16 at 05:35
I am not sure what you are after. HomeID in your example seems to be unique for each element. – akrun Oct 05 '16 at 05:36
I want to know if all columns can be displayed or not – Rohit Venkat Gandhi Mendadhala Oct 05 '16 at 05:40
In that case you may need `mutate_each` instead of `summarise_each`. I am not sure about your expected output – akrun Oct 05 '16 at 05:57

Aggregate concept in R

0 Answers0