I’d like to generate some simple calculations for each combination of two factor variables and store the results in a data frame. Here are the data:

df <- data.frame(SPECIES = as.factor(c(rep("SWAN",10), rep("DUCK",4), rep("GOOS",12),
                             rep("PASS",9), rep("FALC",10))),
                 DRAINAGE = as.factor(c(rep(c("Central", "Upper", "West"),15))),
                 CATCH_QTY = c(1,1,2,5,6,1,2,1,1,1,1,1,3,1,1,2,1,1,2,2,
                               rep(1,25)),
                 TAGGED = c(rep("T",6),NA,"T","T",NA,rep("T",10),NA,NA,
                            rep("T",9),NA,"T","T","T","T",NA,rep("T",7),NA),
                 RECAP = c(rep(NA,6),"RC",NA,NA,"RC",rep(NA,10),"RC","RC",
                            rep(NA,9),"RC",NA,NA,NA,NA,"RC",rep(NA,7),"RC"))

And here is the function:

myfunction <- function(dat, yr, spp, drain){
  dat <- dat %>% filter(SPECIES == spp, DRAINAGE == drain)

    estimatea <<-
      dat %>%
      summarise(NumCaught = sum(CATCH_QTY, na.rm = T),
            NewTags = sum(!is.na(TAGGED)), 
            Recaps = sum(!is.na(RECAP)),
            TotTags = sum(NewTags+Recaps))
    
    dataTest1 <- cbind(yr, spp, drain, estimatea$NumCaught, estimatea$NewTags,
                       estimatea$Recaps, estimatea$TotTags)
}

I’ve mostly been experimenting with nested for loops and have been struggling to store the output in a dataframe given that the variables over which I am iterating are factors, rather than numeric, therefore a lot of the existing answers on stack exchange aren’t relevant. The answer for iterating over factors here
does not show how to store the output.

Some examples of my attempts:

out <- list()

for (i in seq_along(levels(df$SPECIES))) {
  for (j in seq_along(levels(df$DRAINAGE))) {
    out[i,j] <- myfunction(df, "2023", i, j)
  }
}

Error in out[i, j] <- myfunction(df, "2023", i, j) : 
  incorrect number of subscripts on matrix

for (i in seq_along(levels(df$SPECIES))) {
  for (j in seq_along(levels(df$DRAINAGE))) {
    out[i+1,j+1] <- myfunction(df, "2023", i, j)
  }
}

Error in out[i, j] <- myfunction(df, "2023", i, j) : 
  incorrect number of subscripts on matrix

I’ve also considered some non-for loop options, e.g.,

combos <- expand.grid(df$SPECIES, df$DRAINAGE) %>% distinct() %>%
  drop_na() %>% rename(spp = Var1, drain = Var2)
test <- myfunction(df, "2023", combos$spp, combos$drain) #generates incorrect results

sapply(combos$spp, function(x) mapply(myfunction,x,combos$drain))
apply(combos, 2, FUN = myfunction)
Error in UseMethod("filter") : 
  no applicable method for 'filter' applied to an object of class "character"

Ideally, the output dataframe would look something like this:

desired_out <- data.frame(yr = rep("2023",3),
                          spp = c("DUCK", "DUCK", "GOOS"),
                          drain = c("West", "Central", "Upper"),
                          V4 = c(1,3,4),
                          V5 = c(1,1,3),
                          v6 = c(0,0,1),
                          V7 = c(1,1,4))