The Distortionary Effects Of Temporal Aggregation On Granger Causality

Economists often have to use temporally aggregated data in causality tests. A number of theoretical studies have pointed out that temporal aggregation has distorting effects on causal inference. This paper provides a quantitative assessment of the magnitude of the distortions created by temporal aggregation by plugging in theoretical cross covariances into the limiting values of least squares estimates. Some Monte Carlo results and an application are provided to assess the impact in small samples. It is observed that in general the most distorting causal inferences are likely at low levels of temporal aggregation. At high levels of aggregation, causal information concentrates in contemporaneous correlations. At present, a data-based approach is not available to establish the direction of causality between contemporaneously correlated variables.