Arguments
label	type	description
var	variable	data to be ranked
grids	grid set	grid(s) (i.e., independent variables) over which data is to be ranked
Returns
rankvar	variable	same as var except values of rankvar are the ranks of corresponding values of var; rankvar is dependent on the same grids as var

Description

datarank converts the n values along the grid(s) of an input variable to integer ranks from 1 to n along the specified grid(s) in the data set, with 1 corresponding to the largest input value and n corresponding to the smallest value.

Example

SOURCES .NOAA .NCDC .ERSST .version2 .SST
X (-170) (-120) RANGE
Y (-5) (5) RANGE
T (Jan 1971) (Dec 2000) RANGE
[X Y]average
T 12 splitstreamgrid
[T2]datarank

In this example, monthly sea surface temperature values from the ERSST data set are selected from 1971 to 2000 and spatially averaged over the Nino 3.4 region in the equatorial Pacific to produce a single monthly time series for the region. The "T 12 splitstreamgrid" command splits the monthly time grid into two component grids -- "T", which is a 12-point periodic grid that indexes the months of the year, and "T2", which is a 30-point grid that indexes the years 1971 to 2000. [T2] datarank converts the SST values to ranks along the "T2" grid of years. As a result, for each month of the year, SST values from 1971 to 2000 are ranked from largest (1) to smallest (30).

Live Example Link