How to calculate percentiles for many subsets and put in a vector

General
1

I have a question about calculating percentiles for each year in several years. How to do this? A small sample is below, but I have more than 10 years of daily data.

DF = data.frame(
year=c(1980, 1980, 1980, 1980, 1981, 1981, 1981, 1981),
month =c(12, 12, 12, 12, 1, 1, 1, 1),
day= c(28, 29, 30, 31, 1, 2, 3, 4),
value=c(0.60, 0.21, 0.43, 0.44, 0.24, 0.29, 0.21, 0.29))

I want to calculate the 20th percentile of the values in each year. How to do this? I just know the sum, mean, max, and min in aggregate() function. For example, the sum of values in each year is calculated as

value.yr = aggregate(value~year, sum, data=DF)

The 20th percentile in year 1980 is:

value.1980= quantile(subset(DF, year==1980)$value, 0.2)

But if I want to calculate the 20th percentile of the values in each year and put them in a vector, how to adapt the code to something like this?
value.20th = aggregate(value~year, FUN = quantile(0.2), data=DF)
Thanks for any help.

2

Is this what you want to do?

DF = data.frame(
    year=c(1980, 1980, 1980, 1980, 1981, 1981, 1981, 1981),
    month =c(12, 12, 12, 12, 1, 1, 1, 1),
    day= c(28, 29, 30, 31, 1, 2, 3, 4),
    value=c(0.60, 0.21, 0.43, 0.44, 0.24, 0.29, 0.21, 0.29))
library(dplyr)

DF %>% 
    group_by(year) %>% 
    summarise(value.20th = quantile(value, 0.2))
#> # A tibble: 2 x 2
#>    year value.20th
#>   <dbl> <dbl>
#> 1  1980 0.342
#> 2  1981 0.228

Created on 2019-06-11 by the reprex package (v0.3.0)

3

You see a lot of references here to tidy and I hope this solution will encourage you to dig into it.

library(tidyverse)
DF = data.frame(
year=c(1980, 1980, 1980, 1980, 1981, 1981, 1981, 1981),
month =c(12, 12, 12, 12, 1, 1, 1, 1),
day= c(28, 29, 30, 31, 1, 2, 3, 4),
value=c(0.60, 0.21, 0.43, 0.44, 0.24, 0.29, 0.21, 0.29))
> DF %>% group_by(year) %>% summarize(year_20Q = quantile(value, 0.2))
# A tibble: 2 x 2
   year year_20Q
  <dbl>    <dbl>
1  1980    0.342
2  1981    0.228
>
4

Didn't mean to step on your answer to the quantile question. You just beat me to the draw!

Richard

5

Thanks for all your help. It works now. By the way, what is the difference between packages dplyr and tidyverse?

6

dplyr is part of the tidyverse the latest is like an aggregator for a family of packages.

7

If your question's been answered (even if by you), would you mind choosing a solution? (See FAQ below for how).

Having questions checked as resolved makes it a bit easier to navigate the site visually and see which threads still need help.

Thanks

8

This topic was automatically closed 7 days after the last reply. New replies are no longer allowed.