nXplain: Number of EOTs needed for variance explanation

In remote: Empirical Orthogonal Teleconnections in R

Description

The function identifies the number of modes needed to explain a certain amount of variance within the response field.

Usage

## S4 method for signature 'EotStack'
nXplain(x, var = 0.9)

Arguments

x	an EotStack
var	the minimum amount of variance to be explained by the modes

Value

an integer denoting the number of EOTs needed to explain var

Note

This is a post-hoc function. It needs an EotStack created as returned by eot. Depending on the potency of the identified EOTs, it may be necessary to compute a high number of modes in order to be able to explain a large enough part of the variance.

Examples

data(vdendool)

nh_modes <- eot(x = vdendool, y = NULL, n = 3, 
                standardised = FALSE, 
                verbose = TRUE)
             
### How many modes are needed to explain 25% of variance?              
nXplain(nh_modes, 0.25)

remote documentation built on May 1, 2019, 11:30 p.m.