CODEWITHSUNDEEP
raggregate
PJG
PJG

Reputation: 3

Aggregate by specific year in R

Apologies if this question has already been dealt with already on SO, but I cannot seem to find a quick solution as of yet.

I am trying to aggregate a dataset by a specific year. My data frame consists of hourly climate data over a period of 10 years. 

head(df)
#  day month year hour rain temp pressure wind
#1   1     1 2005    0    0  7.6     1016   15
#2   1     1 2005    1    0  8.0     1015   14
#3   1     1 2005    2    0  7.7     1014   15
#4   1     1 2005    3    0  7.8     1013   17
#5   1     1 2005    4    0  7.3     1012   17
#6   1     1 2005    5    0  7.6     1010   17

To calculate daily means from the above dataset, I use this aggregate function 

g <- aggregate(cbind(temp,pressure,wind) ~ day + month + year, d, mean)
options(digits=2)

head(g)
#  day month year temp pressure wind
#1   1     1 2005  6.6     1005   25
#2   2     1 2005  6.5     1018   25
#3   3     1 2005  9.7     1019   22
#4   4     1 2005  7.5     1010   25
#5   5     1 2005  7.3     1008   25
#6   6     1 2005  9.6     1009   26

Unfortunately, I get a huge dataset spanning the whole 10 years (2005 to 2014). I am wondering if anybody would be able to help me tweak the above aggregate code so as I would be able to summaries daily means over a specific year as opposed to all of them in one swipe?

Upvotes: 0

Views: 1028

Answers (2)

Joe
Joe

Reputation: 8611

Dplyr also does it nicely.

library(dplyr)

df %>% 
filter(year==2005) %>% 
group_by(day, month, year) %>% 
summarise_each(funs(mean), temp, pressure, wind)

Upvotes: 0

akrun
akrun

Reputation: 887088

You can use the subset argument in aggregate

aggregate(cbind(temp,pressure,wind) ~ day + month + year, df, 
                     subset=year %in% 2005:2014, mean)

Upvotes: 1

Related Questions