在R中找到n-1个矩阵元素

Question

probably it's a very simple question with a very simple answer but I just can't figure it out by myself. I have a matrix called 'hz' with 1 column and 115 rows (hz[1:115, 1]) and I'm trying to find the values preceding those that are smaller than 1 and replace them. I did the following: hz[c(hz < 1)] , got 11 values, then I tried to find the preceding ones: hz[c(hz < 1) - 1] , expected 11 values but got 114. If I try to find specific elements like hz[c(6, 26, 36)] , and the preceding ones: hz[c(6, 26, 36) - 1] I got 3 values in both cases as expected. So what's the difference? Is it a problem that I have a condition (<1) in the index?

Thank you for your help!

Viktor

Answer 1

Basically you want hz[which(hz < 1) - 1] .

Note, hz < 1 is returning a logical vector, ie, TRUE / FALSE . If you take subtraction: (hz < 1) - 1 , TRUE will be seen as 1 and FALSE will be seen as 0 . Applying which to a logical vector gives you positions (as integers) of TRUE , which is what you want.

Consider the following demonstration:

x <- 1:5

x < 3
#[1]  TRUE  TRUE FALSE FALSE FALSE

(x < 3) - 1
#[1]  0  0 -1 -1 -1

which(x < 3)
#[1] 1 2

which(x < 3) - 1
#[1] 0 1