0

I need a new column of mean grouped by months and the same mean for each month repeated n number of times the month appears.

the grouped mean is calculated but they are unique values, I am unable to make it repeat for n number times side by side for months.

tn_1$GAVG = aggregate(tn_1$FATALITIES, list(tn_1$MONTH), mean)

it gives the following error

 Error in `$<-.data.frame`(`*tmp*`, GAVG, value = list(Group.1 = c("01",  : 
 replacement has 12 rows, data has 6267

the new column must display the mean for each month value.

 structure(list(FATALITIES = c(1L, 2L, 5L, 5L, 3L, 3L, 3L, 4L, 
8L, 1L, 7L, 4L, 3L, 4L, 12L, 4L, 1L, 2L, 3L, 1L, 0L, 0L, 4L, 
1L, 0L, 5L, 0L, 12L, 3L, 2L, 4L, 5L, 1L, 22L, 0L, 1L, 2L, 4L, 
7L, 3L), MONTH = c("04", "04", "04", "04", "05", "05", "05", 
"05", "05", "05", "05", "05", "05", "05", "06", "06", "06", "06", 
"06", "06", "06", "06", "06", "06", "06", "06", "06", "06", "06", 
"06", "06", "06", "06", "06", "06", "06", "06", "06", "06", "06"
)), .Names = c("FATALITIES", "MONTH"), row.names = c(NA, 40L), class = "data.frame")
ambrish dhaka
  • 689
  • 7
  • 27
  • Agree with above ^, a reproducible example would be helpful. You can use `dput()` on your dataframe to give you the code to reproduce your data. A sample of your data should be sufficient – Vivek Katial May 17 '19 at 00:18
  • My workaround has been `mg <- aggregate(tn_1$FATALITIES, list(tn_1$MONTH), mean) names(mg) <- c("MONTH", "GAVG") tn_1 <- left_join(tn_1, mg, by = ("MONTH"))` – ambrish dhaka May 17 '19 at 00:33
  • I need a one liner – ambrish dhaka May 17 '19 at 00:34

2 Answers2

2

Using dplyr:

dat %>% group_by(MONTH) %>% mutate(avg=mean(FATALITIES))
# A tibble: 40 x 3
# Groups:   MONTH [3]
   FATALITIES MONTH   avg
        <int> <chr> <dbl>
 1          1 04     3.25
 2          2 04     3.25
 3          5 04     3.25
 4          5 04     3.25
 5          3 05     4.00
 6          3 05     4.00
 7          3 05     4.00
 8          4 05     4.00
 9          8 05     4.00
10          1 05     4.00
# ... with 30 more rows

Or, if you want to use base R only:

dat$avg<- ave(dat$FATALITIES, dat$MONTH, FUN=mean)

The latter is about 10 times faster:

microbenchmark(one=dat %>% group_by(MONTH) %>% mutate(avg=mean(FATALITIES)), two=ave(dat$FATALITIES, dat$MONTH, FUN=mean))
Unit: microseconds
 expr      min       lq      mean   median        uq      max neval
  one 3698.875 4018.193 4438.3810 4283.864 4650.8455 10019.83   100
  two  265.885  326.586  458.7712  476.840  530.2735   820.31   100
iod
  • 7,412
  • 2
  • 17
  • 36
1

Try with base R ave

df$MEAN=ave(df$FATALITIES,df$MONTH,FUN=mean)
BENY
  • 317,841
  • 20
  • 164
  • 234