Statistical model specification
In regression analysis and related fields such as econometrics, specification refers to the practice of converting a theory into a regression model. This process consists of selecting an appropriate functional form for the model and choosing which variables to include. Model specification is one of the first steps in regression analysis. If an estimated model is misspecified, it will be biased and inconsistent.[1]
Specification error and bias
Specification error occurs in a regression model occurs when an independent variable is correlated with the error term. There may be several different causes behind specification error: An incorrect functional form may have been chosen; the dependent variable may be part of a simultaneous equation (simultaneity bias); an omitted variable may have a relationship with both the dependent variable and one or more of the independent variables (omitted variable bias);[2] an irrelevant variable may be included in the model; there may be measurement error in the independent variables.
The Ramsey reset test can help test for specification error.
References
- ^ Lee, Cheng Few; Lee, John C.; Lee, Alice C.Statistics for Business and Financial Economics. World Scientific Publishing Company. 2nd edition. (December 1999). p 718.
- ^ http://cemood.people.wm.edu/603.html