Detecting causality from observations of dynamic variables.Rosebrock, Daniel.
Abstract: ``Granger causality" is a methodology, first proposed in 1969 by Clive Granger, for detecting causal relationships between two stochastically fluctuating quantities. I will present several equivalent ways of describing this methodology. In Granger's original work, it was assumed that the stochastic quantities are discretized Ornstein-Uhlenbeck processes, that is, AR(1) processes. Thus th... read more
- Tufts University. Department of Mathematics.
- Permanent URL
|To Cite:||DCA Citation Guide|