The analysis of causality between systems is still an important research activity, which finds application in several fields of science. The software presented is a new tool for causality detection and analysis between time series. The proposed technique is based on time-delayed neural networks (TDNN). The tool is developed in MATLAB and it comprises three main functions. The first one returns the total causality between two or more systems of equations. The second tool is used to find the "time horizon", id est the time delay at which the influence between the systems occurs. The last function is a causality feature detection to determine the time intervals, in which the mutual coupling is sufficiently strong to have a real influence on the target.
NetCausality: A time-delayed neural network tool for causality detection and analysis
Murari A;
2021
Abstract
The analysis of causality between systems is still an important research activity, which finds application in several fields of science. The software presented is a new tool for causality detection and analysis between time series. The proposed technique is based on time-delayed neural networks (TDNN). The tool is developed in MATLAB and it comprises three main functions. The first one returns the total causality between two or more systems of equations. The second tool is used to find the "time horizon", id est the time delay at which the influence between the systems occurs. The last function is a causality feature detection to determine the time intervals, in which the mutual coupling is sufficiently strong to have a real influence on the target.I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
