R aggregate by multiple rows and apply summarizing function

Question

I have a data set like this:

  index1 index2       var1       var2
1      1      1 -3.1175877 -0.9776574
2      2      2 -0.6531706 -0.2230984
3      3      1  0.7909643 -0.1172342
4      4      2 -1.8195348 -0.1111046
5      5      1 -0.5132456 -1.1177841
6      6      2 -0.4288158  0.9268719

with more variables than in my example set here. Index1 and Index2 are factors and I would like to apply the mean function to all of them after grouping all rows by index1 and index2. So for each combination of index1 and index2 I want the corresponding mean of the variables.

I have found a lot of similar questions on StackOverflow using aggregate, tapply, by etc but for some reason the very similar examples there don't seem to work for me. For example:

aggregate(vis, by=list(vis$index1, vis$index2), mean)

gives a warning for each row complaining "argument is not numeric or logical: returning NA" and the resulting data frame contains index1 and index2 columns with nothing but NAs.

Any help appreciated :)

Are you looking for `aggregate(. ~ index1 + index2, df, mean)`? (where `df` is your data frame)? — David Arenburg, Nov 16 '14 at 12:28
Or using `data.table` `setDT(df)[, lapply(.SD, mean), by=list(index1, index2)]` — akrun, Nov 16 '14 at 12:31
Or using dplyr package (which is not present in the linked dupe): `library(dplyr) ; df %>% group_by(index1, index2) %>% summarise_each(funs(mean))` — David Arenburg, Nov 16 '14 at 12:35

R aggregate by multiple rows and apply summarizing function

0 Answers0