11
6

Structural causal models for macro-variables in time-series

Abstract

We consider a bivariate time series (Xt,Yt)(X_t,Y_t) that is given by a simple linear autoregressive model. Assuming that the equations describing each variable as a linear combination of past values are considered structural equations, there is a clear meaning of how intervening on one particular XtX_t influences YtY_{t'} at later times t>tt'>t. In the present work, we describe conditions under which one can define a causal model between variables that are coarse-grained in time, thus admitting statements like `setting XX to xx changes YY in a certain way' without referring to specific time instances. We show that particularly simple statements follow in the frequency domain, thus providing meaning to interventions on frequencies.

View on arXiv
Comments on this paper