Search code examples
rmaxfinance

Collecting maximum value from a vector in R

I'm trying to create a function which collects the maximum value from multiple time periods of a population and stores them in a sample of maximum values (ie. Block Maxima)

So far I have: 

function(pop, n)
{
    x <- length(pop)
    sample <- numeric(x/n)
    j=0
    i=0
    while(i<x)
    {
        sample[j] <- max(pop[i:i+n])
        j=j+1
        i=i+n
    }
    return(sample)
}

But, I can not figure out the correct way to calculate the maximum. Can anyone help?

Solution

  • You can use dplyr for that:

    library(dplyr)

block_max <- function(pop, n) {
  data.frame(pop = pop, group = ceiling(seq_along(pop) / n)) %>%
    group_by(group) %>%
    summarise(max_group = max(pop))
}
block_max(1:100, 11)