Skip to contents

This function finds the nearest y to every x. Y's may be duplicated.

Usage

find_nearest(
  x,
  y,
  direction = c("both", "ascending", "descending"),
  returnIndex = FALSE
)

getlower(x, y, upper = FALSE)

Arguments

x

first value

y

second value

direction

default = both, ascending for only y matches before x and descending for only y matches after x.

returnIndex

should an index of the mathced y values be returned instead of the matched list.

upper

upper value?

Value

a list of length 2 with a y matched to every x, note if direction = 'ascending' or 'descending', NAs will be returned for x values with no y values before or after, respectively. OR an index of matched y values if returnIndex = TRUE.

indexes of y for each x

Functions

  • getlower(): function for finding lower(upper) value

References

This function borrowed heavily from this stack exchange post: https://stats.stackexchange.com/questions/161379/quickly-finding-nearest-time-observation