General Considerations and Historiography
Econometrics has evolved into a branch of statistics concerned with the direct testing of models against the data. This is in stark contrast to the natural sciences in which the role of statistical testing has been negligible. Instead, the natural sciences follow a two tier approach. The lower tier involves the determination of empirical regularities, ‘stylized facts’ in the terminology of economics, directly visible in the data. The upper tier is concerned with the theoretical explanation of the stylized facts. Regarding the stylized facts, the aim is to establish them as robust, i.e. evident in diverse and independently generated data sets. This is the meaning of ‘testing’ in natural sciences, as opposed to its meaning in statistics and econometrics. The function of theory is then to provide a simple coherent framework for replicating as many of the stylized facts as possible. The past work of SEMECON has dealt with business cycles at both levels and applied the natural science paradigm to their description and explanation. The first chapters of part II and III of the present work concentrate, respectively, on the description of stylized facts. This is also where most of the narrower methodological issues arise. The present section is therefore mainly concerned with descriptive methodologies.
KeywordsNatural Science Business Cycle Stylize Fact Empirical Regularity Lower Tier
Unable to display preview. Download preview PDF.
- 1.Applications are discussed, for example, in Marple (1980).Google Scholar