Learn R Programming

FluxPoint (version 0.1.2)

Change Point Detection for Non-Stationary and Cross-Correlated Time Series

Description

Implements methods for multiple change point detection in multivariate time series with non-stationary dynamics and cross-correlations. The methodology is based on a model in which each component has a fluctuating mean represented by a random walk with occasional abrupt shifts, combined with a stationary vector autoregressive structure to capture temporal and cross-sectional dependence. The framework is broadly applicable to correlated multivariate sequences in which large, sudden shifts occur in all or subsets of components and are the primary targets of interest, whereas small, smooth fluctuations are not. Although random walks are used as a modeling device, they provide a flexible approximation for a wide class of slowly varying or locally smooth dynamics, enabling robust performance beyond the strict random walk setting.

Copy Link

Version

Install

install.packages('FluxPoint')

Version

0.1.2

License

GPL-2

Maintainer

Yuhan Tian

Last Published

January 10th, 2026

Functions in FluxPoint (0.1.2)

Transform Data for VAR Estimation

Plot multivariate time series with detected change points and estimated means

Randomly generate an autoregressive coefficient matrix \(\Phi\)

Randomly generate an innovation covariance matrix \(\Sigma_{\boldsymbol{\nu}}\)

Matrix square root

Split Coefficient Matrix into VAR Lags

Objective function for robust parameter estimation (RPE)

Cross Validation for Elastic Net VAR Estimation

Cross-Validated VAR Estimation using Elastic Net

Add mean shifts to multivariate time series data

Core change point detection algorithm (given known parameters)

estimatePhinu_nondiag

Estimate non-diagonal VAR(1) parameters after mean removal

estimateCovariance

Estimate Covariance Matrix from Residuals

Apply Thresholding to VAR Coefficients

computeResiduals

Compute VAR Model Residuals

duplicateMatrix

Construct Lagged Design Matrix for VAR

FluxPoint change point detection algorithm

Estimate the fluctuating mean sequence via maximum likelihood

estimate_musseg

Estimate fluctuating mean segmentwise given detected change points

Generate multivariate time series from the proposed model

Evaluate change point detection accuracy metrics

get_Sig_e1_approx

Approximate the long-run covariance matrix \(\Gamma_{\boldsymbol{\epsilon}}(0)\)

Compute the covariance matrix \(\Sigma_{\mathrm{ALL}}^*\) for observations within a moving window

Fit VAR Model with Elastic Net via Cross Validation

estimate_RWVAR_cp

Robust parameter estimation (RPE) for univariate time series

estimate_RWVAR_cp_heter

Robust parameter estimation (RPE) for multivariate time series