Serial Correlation as a Convenient Simplification, not a Nuisance: A Comment on a Study of the Demand for Money by the Bank of England
Oxford University Press eBooks, Oct 26, 2000
Sargan's COMFAC procedure for single equations (common factors in lag polynomials... more Sargan's COMFAC procedure for single equations (common factors in lag polynomials) is exposited. Four important conceptual clarifications arise. First, residual autocorrelation does not entail error autoregression. Secondly, autoregressive errors are common factor dynamics. Thirdly, discriminating between systematic dynamics and error dynamics can only be done in the context of a general‐to‐simple strategy, with autocorrelated errors potentially reducing the parameterization. Fourthly, differencing imposes common factors of unity. The empirical performance of COMFAC was illustrated for modelling broad money demand, and for integrating economic analysis with statistical modelling; long‐run money demand theory guided the empirical estimates of the equilibrium‐correction model.
Uploads
Papers by Grayham Mizon