Data based intervention approach for Complexity-Causality measure

Abstract

Causality testing methods are being widely used in various disciplines of science. Model-free methods for causality estimation are very useful as the underlying model generating the data is often unknown. However, existing model-free measures assume separability of cause and effect at the level of individual samples of measurements and unlike model-based methods do not perform any intervention to learn causal relationships. These measures can thus only capture causality which is by the associational occurrence of ‘cause’ and ‘effect’ between well separated samples. In real-world processes, often ‘cause’ and ‘effect’ are inherently inseparable or become inseparable in the acquired measurements. We propose a novel measure that uses an adaptive interventional scheme to capture causality which is not merely associational. The scheme is based on characterizing complexities associated with the dynamical evolution of processes on short windows of measurements. The formulated measure, Compression- Complexity Causality is rigorously tested on simulated and real datasets and its performance is compared with that of existing measures such as Granger Causality and Transfer Entropy. The proposed measure is robust to presence of noise, long-term memory, filtering and decimation, low temporal resolution (including aliasing), non-uniform sampling, finite length signals and presence of common driving variables. Our measure outperforms existing state-of-the-art measures, establishing itself as an effective tool for causality testing in real world applications.

Understanding dynamical states observed in a thermo-acoustic system from perspective of frequency-phase relationship derived from a probabilistic oscillator model

“…The extension of the same to multivariate data to estimate the level of dependence and subsequently causality complex measure CCC , which we term as causality index in this paper, is defined as [40],…”

“…The information or complexity measure I can be chosen to be any function that represents the dependence of a time-series on the other. [40] employ effort-to-compress (ETC) as the reference method to compute CCC. ETC is based on converting a given time series into symbols and subsequently converting repeatedly occurring pairs of symbols as one given symbol, till the entire time series is compressed and symbolized [41].…”

“…It has been suggested that symbolic analysis based complexity measures are quite robust to noisy and/or coarse data [40].…”

“…Many measures for causality exist [38,39], with varying applicability to the underlying processes. Considering the noisy information contained in stable combustion, we resort to Causality complex measure, an index derived [40]. Causality complexity measure is an information theoretic approach to determine the extent of causality between two dynamic events.…”

Self-resonating Vibrotactile Feedback Instruments ||: Making, Playing, Conceptualising :||

“…[52] , [53]). This provides an analysis tool with which to better understand the experiential affordances of particular parameter settings, material or systemic changes in instrument design, so better informing the iterative design and practice loop.…”