For Loop on ddply

Question

I'm trying to perform a loop with for in R. What I'm trying to do I think is quite simple. I have a vector (but I'm trying also with a column in a data frame) and I have a function in dpply in which R has to substitute the values from the vector:

x = c(1,3,4)

my data frame prova1 is:

   anno variable value
1   1922      gen   0.5
2   1922      gen   0.0
3   1922      gen   1.5
4   1922      gen   0.0
5   1922      gen   4.0
6   1922      gen   2.5
7   1922      gen   5.0
8   1922      gen   0.0
9   1922      gen   0.0
10  1922      gen   0.0
11  1922      gen   0.0
12  1922      gen   0.0
13  1922      gen   0.0
14  1922      gen   0.0
15  1922      gen   0.0
16  1922      gen   2.5
17  1922      gen   0.0
18  1922      gen   0.0

also, I want r to store every result of the function in a list, thus I had created an empty list object

usq<-list()

then I use the loop

  n = length(x)
   for(i in n) {
  usq[[i]] <-ddply(prova1, .(anno),  summarize,
                             sum = sum(value >= x[i] ))
}

the problem is that R overwrite each time the whole list, thus, in the end I obtain a list with the first value NULL and only the last object of the list is correct.

Moreover, I would like to do the same loop on a column or a row like:

   gen   feb   mar   apr   mag   giu   lug   ago   set   ott   nov   dic
  3.93   5.2   3.2     4     5     6     7     8     9    10    11    12

in order to get an object in the list for each column. Is there any way to stop R to overwrite the list? And how to perform the same loop on several column?

Thank you in advance, Luca

Answer 1

Here is one option with dplyr and purrr

library(dplyr)
library(purrr)
map(x, ~ prova1 %>%
            group_by(anno) %>%
            summarise(Sum = sum(value >= .x)))

---

In the OP's loop, 'n' is a single value ie length of 'x' = 3. We are looping over

for(i in 3) # i.e. one time

Instead, it should be

for(i in 1:3)

ie

for(i in seq_len(n))

It is also better to initialize the output list with the predefined length

library(plyr)
usq <- vector('list', length(x))
for(i in seq_len(n)) {
    usq[[i]] <- ddply(prova1, .(anno),  summarize,
                         sum = sum(value >= x[i] ))
    }