How to apply a function on a datafram to take descriptive statistics

Question

I have a list with two dataframe with one variable (Year) that should be factor and another which is numeric and I want its descriptives. This is an example of my list:

> D1
   Year    value
1  1386 7.544808
2  1387 7.552638
3  1387 7.572596
4  1387 7.790549
5  1388 7.607089
6  1388 7.635559
7  1389 7.469881
8  1389 7.622461
9  1389 7.622461
10 1390 7.596479
11 1390 7.645063
12 1391 7.654853
13 1391 7.605891
14 1392 7.612247
15 1381 7.747241
16 1383 7.808759
17 1383 7.834336
18 1384 7.482341
19 1384 7.433035

> D2
   Year    value
1  1386 7.544808
2  1387 7.552638
3  1387 7.572596
4  1387 7.790549
5  1388 7.607089
6  1388 7.635559
7  1389 7.469881
8  1389 7.622461
9  1389 7.622461
10 1390 7.596479
11 1390 7.645063
12 1391 7.654853
13 1391 7.605891
14 1392 7.612247
15 1381 7.747241
16 1383 7.808759
17 1383 7.834336
18 1384 7.482341
19 1384 7.433035

My_list<-list(Labe1=D1,Label2=D2)

Now I want to apply my following function on above list to produce descriptive statistics for variable value for different categories of year.

# take mean with confience interval from columns
MeanFunc<-function(x) round(mean(x,na.rm = TRUE),digits=6 )
SEFunc<-function(x) round(qt(0.975,df=sum(!is.na(x))-1)*sd(x,na.rm = TRUE)/sqrt(sum(!is.na(x)) ),digits=5 )
SDFunc<-function(x) round(sd(x,na.rm = TRUE),digits=5 )
LeftFunc<-function(x)  round(mean(x,na.rm = TRUE)-SEFunc(x),digits=5) 
RightFunc<-function(x) round(mean(x,na.rm = TRUE)+SEFunc(x),digits=5)  
MaxFunc<-function(x) round(max(x,na.rm = TRUE) ,digits=5)  
MinFunc<-function(x) round(min(x,na.rm = TRUE) ,digits=5) 

multi.fun <- function(x) {
  c(Mean = MeanFunc(x), SE = SEFunc(x), SD = SDFunc(x), Left=LeftFunc(x),Right=RightFunc(x),Max=MaxFunc(x),Min=MinFunc(x))
}

Now how can I produce an output as list similar to this?:

$Lable1
Mean      SE      SD    Left   Right     Max     Min
value 7.407750 0.02683 0.35525 7.38092 7.43458 8.54102 5.90301
1381  0.203978 0.09325 1.23486 0.11073 0.29723 8.08833 0.00000
1382  0.078627 0.05813 0.76970 0.02050 0.13676 7.99239 0.00000
1383  0.635951 0.16005 2.11930 0.47590 0.79600 8.54102 0.00000
1384  0.422948 0.13113 1.73636 0.29182 0.55408 8.20205 0.00000
1385  0.267271 0.10543 1.39602 0.16184 0.37270 8.30430 0.00000
1386  0.354070 0.12012 1.59055 0.23395 0.47419 7.85514 0.00000
1387  1.279604 0.21165 2.80268 1.06795 1.49125 8.23982 0.00000
$Lable2
Mean      SE      SD    Left   Right     Max     Min
value 7.407750 0.02683 0.35525 7.38092 7.43458 8.54102 5.90301
1381  0.203978 0.09325 1.23486 0.11073 0.29723 8.08833 0.00000
1382  0.078627 0.05813 0.76970 0.02050 0.13676 7.99239 0.00000
1383  0.635951 0.16005 2.11930 0.47590 0.79600 8.54102 0.00000
1384  0.422948 0.13113 1.73636 0.29182 0.55408 8.20205 0.00000
1385  0.267271 0.10543 1.39602 0.16184 0.37270 8.30430 0.00000
1386  0.354070 0.12012 1.59055 0.23395 0.47419 7.85514 0.00000
1387  1.279604 0.21165 2.80268 1.06795 1.49125 8.23982 0.00000

Thanks a lot...

something like `lapply(dfList, function(x) aggregate(value ~ year, data=x, FUN=multi.func))` where dfList is a list containing your data.frames. See my answer [here](https://stackoverflow.com/questions/17499013/how-do-i-make-a-list-of-data-frames) for how to construct a list of data.frames if they aren't in a list initially. — lmo, Jan 26 '19 at 13:35

score 0 · Answer 1 · answered Jan 26 '19 at 14:12

Check this solution:

library(tidyverse)
library(plotrix)

My_list %>%
  map(
    ~group_by(.x, Year) %>% 
    summarise(
      Mean = mean(value, na.rm = TRUE) %>% round(6),
      SE = std.error(value, na.rm = TRUE),
      SD = sd(value,na.rm = TRUE),
      Left = mean(value,na.rm = TRUE) - std.error(value, na.rm = TRUE),
      Right = mean(value,na.rm = TRUE) + std.error(value, na.rm = TRUE),
      Max = max(value,na.rm = TRUE),
      Min = min(value,na.rm = TRUE)
    ) %>%
      mutate_at(3:8, ~round(.x, 5))
  )

How to apply a function on a datafram to take descriptive statistics

1 Answers1