lowerBound in weighted.quantile function

[javakh]

You have used lowerBound function inside the script of weighted.quantile function.
[weighted.quantile <- function (x, w, probs = seq(0, 1, 0.25), na.rm = FALSE) {
stopifnot(length(x) == length(w))
na <- is.na(x) | is.na(w)
if (any(na)) {
if (na.rm) {
x <- x[!na]
w <- w[!na]
} else {
stop("Missing values not allowed when na.rm is FALSE", call. = FALSE)
}
}

Ensure x and w in ascending order of x

ord <- order(x)
x <- x[ord]
w <- w[ord]

Find closest x just below and above index

n <- sum(w)
index <- 1 + (n - 1) * probs
j <- floor(index)

wts <- cumsum(w)
lo <- x[lowerBound(j, wts)] # X_j
hi <- x[lowerBound(j + 1, wts)]

g <- index - j
ifelse(lo == hi, lo, (1 - g) * lo + g * hi)
}
]

But the lowerBound function does not exist in R.

?lowerBound
No documentation for ‘lowerBound’ in specified packages and libraries:
you could try ‘??lowerBound

??lowerBound also showed there is no such function in R.

So, How does it run inside your script???

[gvegayon]

Here you have

bigvis/src/lowerBound.cpp

Line 8 in 9cce240

IntegerVector lowerBound(const NumericVector& x, const NumericVector& breaks) {