Learning Multi-Frequency Partial Correlation Graphs
This addresses a problem for researchers and practitioners in time series analysis, particularly in fields like finance, by enabling frequency-specific dependency insights, though it is incremental as it builds on existing partial correlation methods.
The paper tackled the limitation of existing methods that learn partial correlations in time series but fail to discriminate across frequency bands, by proposing methods to learn frequency-dependent partial correlation graphs, with results showing outperformance over state-of-the-art on synthetic data and confirmation of partial correlations existing only in a few frequency bands in financial data.
Despite the large research effort devoted to learning dependencies between time series, the state of the art still faces a major limitation: existing methods learn partial correlations but fail to discriminate across distinct frequency bands. Motivated by many applications in which this differentiation is pivotal, we overcome this limitation by learning a block-sparse, frequency-dependent, partial correlation graph, in which layers correspond to different frequency bands, and partial correlations can occur over just a few layers. To this aim, we formulate and solve two nonconvex learning problems: the first has a closed-form solution and is suitable when there is prior knowledge about the number of partial correlations; the second hinges on an iterative solution based on successive convex approximation, and is effective for the general case where no prior knowledge is available. Numerical results on synthetic data show that the proposed methods outperform the current state of the art. Finally, the analysis of financial time series confirms that partial correlations exist only within a few frequency bands, underscoring how our methods enable the gaining of valuable insights that would be undetected without discriminating along the frequency domain.