Causality Testing With Messy Data: Some Preliminary Experimental Evidence
Year: 1984
Type: preprint
Abstract: Two tests for bivariate causal ordering. are examined using nonstationary data series with known causal structures. Instantaneous causality is often identified when not present due to the presence of the nonstationary components. Increased degrees of error covariance between the series leads to incorrect conclusions and collinearity problems in the tests.
Authors: Gamber, Edward N., Hudson, Michael A.
Language: English
Domain: Social Sciences
Open Access status: green
OpenAlex ID: W2945029867
