Abstract:Identifying directed spectral information flow between multivariate time series is important for many applications in finance, climate, geophysics and neuroscience. Spectral Granger causality (SGC) is a prediction-based measure characterizing directed information flow at specific oscillatory frequencies. However, traditional vector autoregressive (VAR) approaches are insufficient to assess SGC when time series have mixed frequencies (MF) or are coupled by nonlinearity. Here we propose a time-frequency canonical correlation analysis approach ("MF-TFCCA") to assess the strength and driving frequency of spectral information flow. We validate the approach with intensive computer simulations on MF time series under various interaction conditions and assess statistical significance of the estimate with surrogate data. We further apply MF-TFCCA to real-life finance, climate and neuroscience data. Our analysis framework provides an exploratory and computationally efficient approach to quantify directed information flow between MF time series in the presence of complex and nonlinear interactions.

What problem does this paper attempt to address?

The problem that this paper attempts to solve is: **How to infer directed spectral information flow between time series of mixed frequencies?** Specifically, the traditional vector autoregressive (VAR) method has deficiencies when dealing with time series with different sampling frequencies or non - linearly coupled, and it cannot accurately evaluate Spectral Granger Causality (SGC). Therefore, the authors propose a new method - **Mixed - Frequency Time - Frequency Canonical Correlation Analysis (MF - TFCCA)** to overcome these limitations. ### Background and challenges of the problem 1. **Directed information flow in multivariate time series**: - In many fields (such as finance, climate, geophysics, and neuroscience), it is very important to identify the directed information flow between multivariate time series. - Spectral Granger Causality (SGC) is a prediction - based measure used to characterize the directed information flow at a specific oscillation frequency. 2. **Limitations of traditional methods**: - Traditional VAR methods perform poorly when dealing with Mixed - Frequency (MF) time series or non - linear couplings. - When time series have different sampling frequencies, traditional GC methods cannot work effectively, especially when there are large differences in sampling frequencies or non - linear interactions exist. ### Proposed method and its advantages 1. **MF - TFCCA method**: - **Adapt to mixed frequencies**: MF - TFCCA can evaluate the directed spectral information flow between time series with different sampling frequencies. - **Non - parametric method**: This method is non - parametric and does not need to assume a specific distribution form of the data. - **High computational efficiency**: Compared with the traditional MF - VAR method, MF - TFCCA has significant improvements in computational efficiency and detection accuracy. 2. **Verification and application**: - The effectiveness of MF - TFCCA has been verified through intensive computer simulations. - It has been applied to actual financial, climate, and neuroscience data, demonstrating its applicability in complex and non - linear interaction cases. ### Specific application scenarios - **Financial field**: For example, comparing the information flow between daily stock prices and weekly trading volumes. - **Climate forecasting**: For example, relating monthly rainfall at the city level to annual rainfall at the national level. - **Neuroscience**: For example, simultaneously recording neural signals of different modalities (such as calcium imaging and electrophysiology) and studying the directed information flow between them. ### Summary The core problem of the paper is to develop a new method (MF - TFCCA) to solve the problem of directed spectral information flow between time series of mixed frequencies, especially in the presence of complex and non - linear interactions. This method not only improves computational efficiency and detection accuracy but also provides a new tool for multivariate time series analysis across fields.

Inferring directed spectral information flow between mixed-frequency time series

Measuring Causality by Taking the Directional Symbolic Mutual Information Approach

Causality analysis of neural connectivity: critical examination of existing methods and advances of new methods.

Spatio-temporal Granger causality: A new framework.

A method to assess Granger causality, isolation and autonomy in the time and frequency domains: theory and application to cerebrovascular variability

Multiscale Granger causality

Causality between time series

FIGCI: Flow-Based Information-Geometric Causal Inference

Statistical Inference for Local Granger Causality

A framework for assessing frequency domain causality in physiological time series with instantaneous effects

Inferring Weighted Directed Association Network from Multivariate Time Series with a Synthetic Method of Partial Symbolic Transfer Entropy Spectrum and Granger Causality

SCAU: Modeling spectral causality for multivariate time series with applications to electroencephalograms

Continuum - removed absorption features estimate tropical savanna grass quality in situ

Is Granger Causality a Viable Technique for Analyzing fMRI Data?

Dimension Reduction of Polynomial Regression Models for the Estimation of Granger Causality in High-Dimensional Time Series

Causal Inference From Slowly Varying Nonstationary Processes

Multifractal temporally weighted detrended partial cross-correlation analysis to quantify intrinsic power-law cross-correlation of two non-stationary time series affected by common external factors

A feature-based information-theoretic approach for detecting interpretable, long-timescale pairwise interactions from time series

Flow Spatiotemporal Moran's I: Measuring the Spatiotemporal Autocorrelation of Flow Data

Unraveling the Cause-Effect Relation Between Time Series.

Spectral Time-Varying Pattern Causality and Its Application