Work on lists of datasets instead of individual datasets by using functional programming

December 20, 2016
By

(This article was first published on Econometrics and Free Software, and kindly contributed to R-bloggers)


Analyzing a lot of datasets can be tedious. In my work, I often have to compute descriptive statistics, or plot some graphs for some variables for a lot of datasets. The variables in question have the same name accross the datasets but are measured for different years. As an example, imagine you have this situation:

data2000 <- mtcars
data2001 <- mtcars

For the sake of argument, imagine that data2000 is data from a survey conducted in the year 2000 and data2001 is the same survey but conducted in the year 2001. For illustration purposes, I use the mtcars dataset, but I could have used any other example. In these sort of situations, the variables are named the same in both datasets. Now if I want to check the summary statistics of a variable, I might do it by running:

summary(data2000$cyl)
summary(data2001$cyl)

but this can get quite tedious, especially if instead of only having two years of data, you have 20 years. Another possibility is to merge both datasets and then check the summary statistics of the variable of interest. But this might require a lot of preprocessing, and sometimes you really just want to do a quick check, or some dirty graphs. So you might be tempted to write a loop, which would require to put these two datasets in some kind of structure, such as a list:

list_data <- list("data2000" = data2000, "data2001" = data2001)

for (i in 1:2){
    print(summary(list_data[[i]]$cyl))
 }
##    Min. 1st Qu.  Median    Mean 3rd Qu.    Max. 
##   4.000   4.000   6.000   6.188   8.000   8.000 
##    Min. 1st Qu.  Median    Mean 3rd Qu.    Max. 
##   4.000   4.000   6.000   6.188   8.000   8.000

But this also might get tedious, especially if you want to do this for a lot of different variables, and want to use different functions than summary().

Another, simpler way of doing this, is to use purrr::map() or lapply(). But there is a catch though: how do we specify the column we want to work on? Let’s try some things out:

library(purrr)

map(list_data, summary(cyl))
Error in summary(cyl) : object 'cyl' not found

Maybe this will work:

map(list_data, summary, cyl)
## $data2000
##       mpg             cyl             disp             hp       
##  Min.   :10.40   Min.   :4.000   Min.   : 71.1   Min.   : 52.0  
##  1st Qu.:15.43   1st Qu.:4.000   1st Qu.:120.8   1st Qu.: 96.5  
##  Median :19.20   Median :6.000   Median :196.3   Median :123.0  
##  Mean   :20.09   Mean   :6.188   Mean   :230.7   Mean   :146.7  
##  3rd Qu.:22.80   3rd Qu.:8.000   3rd Qu.:326.0   3rd Qu.:180.0  
##  Max.   :33.90   Max.   :8.000   Max.   :472.0   Max.   :335.0  
##       drat             wt             qsec             vs        
##  Min.   :2.760   Min.   :1.513   Min.   :14.50   Min.   :0.0000  
##  1st Qu.:3.080   1st Qu.:2.581   1st Qu.:16.89   1st Qu.:0.0000  
##  Median :3.695   Median :3.325   Median :17.71   Median :0.0000  
##  Mean   :3.597   Mean   :3.217   Mean   :17.85   Mean   :0.4375  
##  3rd Qu.:3.920   3rd Qu.:3.610   3rd Qu.:18.90   3rd Qu.:1.0000  
##  Max.   :4.930   Max.   :5.424   Max.   :22.90   Max.   :1.0000  
##        am              gear            carb      
##  Min.   :0.0000   Min.   :3.000   Min.   :1.000  
##  1st Qu.:0.0000   1st Qu.:3.000   1st Qu.:2.000  
##  Median :0.0000   Median :4.000   Median :2.000  
##  Mean   :0.4062   Mean   :3.688   Mean   :2.812  
##  3rd Qu.:1.0000   3rd Qu.:4.000   3rd Qu.:4.000  
##  Max.   :1.0000   Max.   :5.000   Max.   :8.000  
## 
## $data2001
##       mpg             cyl             disp             hp       
##  Min.   :10.40   Min.   :4.000   Min.   : 71.1   Min.   : 52.0  
##  1st Qu.:15.43   1st Qu.:4.000   1st Qu.:120.8   1st Qu.: 96.5  
##  Median :19.20   Median :6.000   Median :196.3   Median :123.0  
##  Mean   :20.09   Mean   :6.188   Mean   :230.7   Mean   :146.7  
##  3rd Qu.:22.80   3rd Qu.:8.000   3rd Qu.:326.0   3rd Qu.:180.0  
##  Max.   :33.90   Max.   :8.000   Max.   :472.0   Max.   :335.0  
##       drat             wt             qsec             vs        
##  Min.   :2.760   Min.   :1.513   Min.   :14.50   Min.   :0.0000  
##  1st Qu.:3.080   1st Qu.:2.581   1st Qu.:16.89   1st Qu.:0.0000  
##  Median :3.695   Median :3.325   Median :17.71   Median :0.0000  
##  Mean   :3.597   Mean   :3.217   Mean   :17.85   Mean   :0.4375  
##  3rd Qu.:3.920   3rd Qu.:3.610   3rd Qu.:18.90   3rd Qu.:1.0000  
##  Max.   :4.930   Max.   :5.424   Max.   :22.90   Max.   :1.0000  
##        am              gear            carb      
##  Min.   :0.0000   Min.   :3.000   Min.   :1.000  
##  1st Qu.:0.0000   1st Qu.:3.000   1st Qu.:2.000  
##  Median :0.0000   Median :4.000   Median :2.000  
##  Mean   :0.4062   Mean   :3.688   Mean   :2.812  
##  3rd Qu.:1.0000   3rd Qu.:4.000   3rd Qu.:4.000  
##  Max.   :1.0000   Max.   :5.000   Max.   :8.000

Not quite! You get the summary statistics of every variable, cyl simply gets ignored. This might be ok in our small toy example, but if you have dozens of datasets with hundreds of variables, the output becomes unreadable. The solution is to use an anonymous functions:

map(list_data, (function(x) summary(x$cyl)))
## $data2000
##    Min. 1st Qu.  Median    Mean 3rd Qu.    Max. 
##   4.000   4.000   6.000   6.188   8.000   8.000 
## 
## $data2001
##    Min. 1st Qu.  Median    Mean 3rd Qu.    Max. 
##   4.000   4.000   6.000   6.188   8.000   8.000

This is, in my opinion, much more readable than a loop, and the output of this is another list, so it’s easy to save it:

summary_cyl <- map(list_data, (function(x) summary(x$cyl)))
str(summary_cyl)
## List of 2
##  $ data2000:Classes 'summaryDefault', 'table'  Named num [1:6] 4 4 6 6.19 8 ...
##   .. ..- attr(*, "names")= chr [1:6] "Min." "1st Qu." "Median" "Mean" ...
##  $ data2001:Classes 'summaryDefault', 'table'  Named num [1:6] 4 4 6 6.19 8 ...
##   .. ..- attr(*, "names")= chr [1:6] "Min." "1st Qu." "Median" "Mean" ...

With the loop, you would need to “allocate” an empty list that you would fill at each iteration.

So this is already nice, but wouldn’t it be nicer to simply have to type:

summary(list_data$cyl)

and have the summary of variable cyl for each dataset in the list? Well it is possible with the following function I wrote to make my life easier:

to_map <- function(func){
  function(list, column, ...){
    if(missing(column)){
        res <- purrr::map(list, (function(x) func(x, ...)))
      } else {
        res <- purrr::map(list, (function(x) func(x[column], ...)))
             }
    res
  }
}

By following this chapter of Hadley Wickham’s book, Advanced R, I was able to write this function. What does it do? It basically generalizes a function to work on a list of datasets instead of just on a dataset. So for example, in the case of summary():

summarymap <- to_map(summary)

summarymap(list_data, "cyl")
## $data2000
##       cyl       
##  Min.   :4.000  
##  1st Qu.:4.000  
##  Median :6.000  
##  Mean   :6.188  
##  3rd Qu.:8.000  
##  Max.   :8.000  
## 
## $data2001
##       cyl       
##  Min.   :4.000  
##  1st Qu.:4.000  
##  Median :6.000  
##  Mean   :6.188  
##  3rd Qu.:8.000  
##  Max.   :8.000

So now everytime I want to have summary statistics for a variable, I just need to use summarymap():

summarymap(list_data, "mpg")
## $data2000
##       mpg       
##  Min.   :10.40  
##  1st Qu.:15.43  
##  Median :19.20  
##  Mean   :20.09  
##  3rd Qu.:22.80  
##  Max.   :33.90  
## 
## $data2001
##       mpg       
##  Min.   :10.40  
##  1st Qu.:15.43  
##  Median :19.20  
##  Mean   :20.09  
##  3rd Qu.:22.80  
##  Max.   :33.90

If I want the summary statistics for every variable, I simply omit the column name:

summarymap(list_data)
## $data2000
##       mpg             cyl             disp             hp       
##  Min.   :10.40   Min.   :4.000   Min.   : 71.1   Min.   : 52.0  
##  1st Qu.:15.43   1st Qu.:4.000   1st Qu.:120.8   1st Qu.: 96.5  
##  Median :19.20   Median :6.000   Median :196.3   Median :123.0  
##  Mean   :20.09   Mean   :6.188   Mean   :230.7   Mean   :146.7  
##  3rd Qu.:22.80   3rd Qu.:8.000   3rd Qu.:326.0   3rd Qu.:180.0  
##  Max.   :33.90   Max.   :8.000   Max.   :472.0   Max.   :335.0  
##       drat             wt             qsec             vs        
##  Min.   :2.760   Min.   :1.513   Min.   :14.50   Min.   :0.0000  
##  1st Qu.:3.080   1st Qu.:2.581   1st Qu.:16.89   1st Qu.:0.0000  
##  Median :3.695   Median :3.325   Median :17.71   Median :0.0000  
##  Mean   :3.597   Mean   :3.217   Mean   :17.85   Mean   :0.4375  
##  3rd Qu.:3.920   3rd Qu.:3.610   3rd Qu.:18.90   3rd Qu.:1.0000  
##  Max.   :4.930   Max.   :5.424   Max.   :22.90   Max.   :1.0000  
##        am              gear            carb      
##  Min.   :0.0000   Min.   :3.000   Min.   :1.000  
##  1st Qu.:0.0000   1st Qu.:3.000   1st Qu.:2.000  
##  Median :0.0000   Median :4.000   Median :2.000  
##  Mean   :0.4062   Mean   :3.688   Mean   :2.812  
##  3rd Qu.:1.0000   3rd Qu.:4.000   3rd Qu.:4.000  
##  Max.   :1.0000   Max.   :5.000   Max.   :8.000  
## 
## $data2001
##       mpg             cyl             disp             hp       
##  Min.   :10.40   Min.   :4.000   Min.   : 71.1   Min.   : 52.0  
##  1st Qu.:15.43   1st Qu.:4.000   1st Qu.:120.8   1st Qu.: 96.5  
##  Median :19.20   Median :6.000   Median :196.3   Median :123.0  
##  Mean   :20.09   Mean   :6.188   Mean   :230.7   Mean   :146.7  
##  3rd Qu.:22.80   3rd Qu.:8.000   3rd Qu.:326.0   3rd Qu.:180.0  
##  Max.   :33.90   Max.   :8.000   Max.   :472.0   Max.   :335.0  
##       drat             wt             qsec             vs        
##  Min.   :2.760   Min.   :1.513   Min.   :14.50   Min.   :0.0000  
##  1st Qu.:3.080   1st Qu.:2.581   1st Qu.:16.89   1st Qu.:0.0000  
##  Median :3.695   Median :3.325   Median :17.71   Median :0.0000  
##  Mean   :3.597   Mean   :3.217   Mean   :17.85   Mean   :0.4375  
##  3rd Qu.:3.920   3rd Qu.:3.610   3rd Qu.:18.90   3rd Qu.:1.0000  
##  Max.   :4.930   Max.   :5.424   Max.   :22.90   Max.   :1.0000  
##        am              gear            carb      
##  Min.   :0.0000   Min.   :3.000   Min.   :1.000  
##  1st Qu.:0.0000   1st Qu.:3.000   1st Qu.:2.000  
##  Median :0.0000   Median :4.000   Median :2.000  
##  Mean   :0.4062   Mean   :3.688   Mean   :2.812  
##  3rd Qu.:1.0000   3rd Qu.:4.000   3rd Qu.:4.000  
##  Max.   :1.0000   Max.   :5.000   Max.   :8.000

I can use any function:

tablemap <- to_map(table)

tablemap(list_data, "cyl")
## $data2000
## 
##  4  6  8 
## 11  7 14 
## 
## $data2001
## 
##  4  6  8 
## 11  7 14
tablemap(list_data, "mpg")
## $data2000
## 
## 10.4 13.3 14.3 14.7   15 15.2 15.5 15.8 16.4 17.3 17.8 18.1 18.7 19.2 19.7 
##    2    1    1    1    1    2    1    1    1    1    1    1    1    2    1 
##   21 21.4 21.5 22.8 24.4   26 27.3 30.4 32.4 33.9 
##    2    2    1    2    1    1    1    2    1    1 
## 
## $data2001
## 
## 10.4 13.3 14.3 14.7   15 15.2 15.5 15.8 16.4 17.3 17.8 18.1 18.7 19.2 19.7 
##    2    1    1    1    1    2    1    1    1    1    1    1    1    2    1 
##   21 21.4 21.5 22.8 24.4   26 27.3 30.4 32.4 33.9 
##    2    2    1    2    1    1    1    2    1    1

I hope you will find this little function useful, and as usual, for any comments just drop me an email by clicking the red enveloppe in the top right corner or tweet me.

To leave a comment for the author, please follow the link and comment on their blog: Econometrics and Free Software.

R-bloggers.com offers daily e-mail updates about R news and tutorials on topics such as: Data science, Big Data, R jobs, visualization (ggplot2, Boxplots, maps, animation), programming (RStudio, Sweave, LaTeX, SQL, Eclipse, git, hadoop, Web Scraping) statistics (regression, PCA, time series, trading) and more...



If you got this far, why not subscribe for updates from the site? Choose your flavor: e-mail, twitter, RSS, or facebook...

Comments are closed.

Sponsors

Never miss an update!
Subscribe to R-bloggers to receive
e-mails with the latest R posts.
(You will not see this message again.)

Click here to close (This popup will not appear again)