Group data in data frame based on date range in second data frame In R

Question

I have the following two data frames:

Date <- seq(as.Date("2013/1/1"), by = "day", length.out = 46)
x <-data.frame(Date)
x$discharge <- c("1000","1100","1200","1300","1400","1200","1300","1300","1200","1100","1200","1200","1100","1400","1200","1100","1400","1000","1100","1200","1300","1400","1200","1300","1300","1200","1100","1200","1200","1100","1400","1200","1100","1400","1000","1100","1200","1300","1400","1200","1300","1300","1200","1100","1200","1200")
x$discharge <- as.numeric(x$discharge)

And

Date_from <- c("2013-01-01","2013-01-15","2013-01-21","2013-02-10")
Date_to <- c("2013-01-07","2013-01-20","2013-01-25","2013-02-15")
y <- data.frame(Date_from,Date_to)
y$concentration <- c("1.5","2.5","1.5","3.5")
y$Date_from <- as.Date(y$Date_from)
y$Date_to <- as.Date(y$Date_to)
y$concentration <- as.numeric(y$concentration)

I am trying to calculate the average discharge from the daily discharges in data frame x for each row in data frame y based on the date range Date_from to Date_to in data frame y . Notice, that there is a gap in the measurements in data frame y between 2013-01-08 to 2013-01-14, and 2013-01-26 to 2013-02-09. This gap is due to the fact that no measurements were taken during this time. And this gap is causing me headaches as I was using the following code to calculate the average discharge for each date range in y :

rng <- cut(x$Date, breaks=c(y$Date_from, max(y$Date_to), 
                    include.lowest=T))
range<-cbind(x,rng)
discharge<-aggregate(cbind(mean=x$discharge)~rng, FUN=mean)

However, if you check the range in data frame range the range for 2013-01-01 to 2013-01-07 is extended up to 2013-01-14 but I only need it to 2013-01-07 and than with a break until the next range begins on 2013-01-15.

Answer 1

You can try a tidyverse .

library(tidyverse)
y %>% 
  split(seq_along(1:nrow(.))) %>% 
  map(~filter(x, between(Date, .$Date_from, .$Date_to)) %>% 
        summarise(Mean=mean(discharge))) %>% 
  bind_rows() %>% 
  bind_cols(y,.)
   Date_from    Date_to concentration     Mean
1 2013-01-01 2013-01-07           1.5 1214.286
2 2013-01-15 2013-01-20           2.5 1166.667
3 2013-01-21 2013-01-25           1.5 1300.000
4 2013-02-10 2013-02-15           3.5 1216.667

Using only this code you can see the values and groups.

y %>% 
  split(seq_along(1:nrow(.))) %>% 
  map(~filter(x, between(Date, .$Date_from, .$Date_to))) 

Answer 2

Here's a base answer:

helper <- merge(x, y)
helper <- helper[helper$Date >= helper$Date_from & helper$Date <= helper$Date_to, ]
aggregate(helper$discharge,
          list(Date_from = helper$Date_from,
               Date_to = helper$Date_to),
          FUN = 'mean')

   Date_from    Date_to        x
1 2013-01-01 2013-01-07 1214.286
2 2013-01-15 2013-01-20 1166.667
3 2013-01-21 2013-01-25 1300.000
4 2013-02-10 2013-02-15 1216.667