gen_dat_lin: Generate Sample Data

In dpweix/mlmcusum: Detects Changes in Covariance for Multivariate TS

Generate Sample Data

Description

Generates a three dimensional multivariate time series for use in simulation studies. Multivariate time series data are generated with either linear or non-linear relationships between variables. Further, the option to produce nonstationary data by adding a sine wave is included. \n Each time a multivariate time series is generated four data frames are produced. There are three different faults that are be introduced, and one data frame in which no fault is introduced.

Usage

gen_dat_lin(n_ic = 1500, n_oc = 500, phi = 0.8, a_1 = 3, a_2 = 2)

gen_dat_ltl(
  n_ic = 1500,
  n_oc = 500,
  phi = 0.8,
  sin_scale = 1.5,
  a_1 = 3,
  a_2 = 2
)

gen_dat_nlr(n_ic = 1500, n_oc = 500, phi = 0.8, a_1 = 3, a_2 = 2)

gen_dat_ltm(
  n_ic = 1500,
  n_oc = 500,
  phi = 0.8,
  sin_scale = 1.5,
  a_1 = 3,
  a_2 = 2
)

Arguments

n_ic	The number of observations which are in-control.
n_oc	The number of observations which are out-of-control, i.e. faulty.
phi	The strength of the autocorrelation in the base time series
a_1	Increase in variance of x2 for fault 1.
a_2	Increase in variance of x1, x2, and x3 for fault 3.
sin_scale	The scale of the sine wave for non-stationarity.

Value

A list of four data frames where the first n_ic observations are identical, but the last n_oc observations have experienced fault 1, 2, 3, or no fault.

dpweix/mlmcusum documentation built on July 31, 2023, 10:13 a.m.