crossing_rate: Multivariate test

In mkoohafkan/wqptools: Outlier Detection Working Group Toolbox

Description

Perform a multivariate rate of change test. This function can also be used for the neighbor test. See https://cdn.ioos.noaa.gov/media/2017/12/qartod_temperature_salinity_manual.pdf for more information.

Usage

crossing_rate(x, y, n.dev, n.prior)

Arguments

x	A vector of values.
y	A vector of coincident observations, e.g. from a separate sensor or nearby station.
n.dev	The number of standard deviations to test against.
n.prior	the number of prior observations to use for computing the standard deviation. For example, to compute standard deviations over the previous 25 hours from 15-minute data use n.prior = 100.

Examples

fake.data1 = c(rnorm(10,10), rnorm(10, 50), rnorm(10, 10))
fake.data2 = rep(10,30)
crossing_rate(fake.data1, fake.data2, c(1.5, 3), 5)
crossing_rate(fake.data2, fake.data1, c(1.5, 3), 5)
crossing_rate(fake.data1, fake.data1, 1.5, 5)

mkoohafkan/wqptools documentation built on May 2, 2021, 8:12 p.m.