Simplicity, Scientific Inference and Econometric Modelling

On Standard-Error-Decreasing Complementarity: Why Collinearity is Not the Whole Story

“…Adding a variable to the model or subtracting one from it may change the estimated effects of the remaining variables due to omitted variable biases and/or estimation efficiency because of changes in standard errors. If smaller, more efficiently estimated models are desired as a basis for the interpretation of results (for arguments in favour of that, see, e.g., Hendry (2007), Keuzenkamp and McAleer (1995), and Hayo (1998)), then these should be derived via a consistent reduction of the general model that includes all theoretically relevant variables. Researchers should report the general model and the corresponding reduction test that supports working with a simplified model.…”

Say's Law: A Rigorous Restatement

“…Randomness may turn out to be a superficial explanation. However, the chief merits of the random hypothesis are simplicity and testability, therefore it has to be applied first -for good methodological reasons (Kreuzenkamp and McAleer, 1995).…”

Econometric Methodology I: The Role of Theory and Observation in Agricultural Economics Research / Ekonometriese Metodologiei: Die Rol Van Teorie en Waarneming in Landbou-Ekonomiese Navorsing

“…Marginalising and conditioning are key notions of this theory of reduction (Keuzenkamp & McAleer, 1995). These models are a simplified representation intended to capture the salient features of some group of phenomena.…”