Stylized fact

Jan Tinbergen When describing what is generally regarded as the first econometric macro model ever developed, Jan Tinbergen (1936) introduced the concept of stylization as follows: Nicholas Kaldor The term "stylised facts" was introduced by the economist Nicholas Kaldor in the context of a debate on economic growth theory in 1961, expanding on model assumptions made in a 1957 paper. Criticizing the neoclassical models of economic growth of his time, Kaldor argued that theory construction should begin with a summary of the relevant facts. However, to handle the problem that "facts as recorded by statisticians, are always subject to numerous snags and qualifications, and for that reason are incapable of being summarized", he suggested that theorists "should be free to start off with a stylised view of the facts – i.e. concentrate on broad tendencies, ignoring individual detail". With respect to broad tendencies that result from such a process, Kaldor coined the term "stylized facts". ==Examples==

Stylized facts are widely used in economics, in particular to motivate the construction of a model and/or to validate it. Examples are: • Stock returns are uncorrelated and not easily forecastable. • Yield curves tend to move in parallel. • Education is positively correlated to lifetime earnings • Inventory behavior of firms: "the variance of production exceeds the variance of sales" ==Uses==

Already in the original paper, Kaldor used his stylized facts of economic growth to argue in favor of his suggested model in comparison to older neoclassical models of economic growth. This idea has been highlighted subsequently by Boland, that the advantages of one model over the other can be set in a clear perspective via the reference of the stylized facts the respective models can explain. Additionally, stylized facts can be used to look "under the hood of models", i.e. be used to validate assumptions of model in a focused way. This can be of particular importance in more complex models. Econophysics/Statistical finance begins from stylized facts. Furthermore, the considerable potential of Stylized Facts for Information Systems research has been investigated and discussed in recent years. ==Reception by other economists==

In an early response, Robert Solow pinpointed a possible problem of stylized facts, stating "there is no doubt that they are stylized, though it is possible to question whether they are facts". The criticized practice of deriving stylized facts ad hoc is still quite prevalent in economics. Still, some possible approaches to derive stylized facts from empirical evidence have been suggested, such as surveying experts, statistically analysing large data sets (especially suitable for financial markets) or aggregating both qualitative and quantitative data from different empirical methods by following a systematic process. == See also ==