General-to-specific Modeling: An Overview and Selected Bibliography
read more
Citations
Promoting novelty, rigor, and style in energy social science: towards codes of practice for appropriate methods and research design
Forecasting and operational research : a review
Mortality Density Forecasts: An Analysis of Six Stochastic Mortality Models
What Segments Equity Markets
Mortality density forecasts: An analysis of six stochastic mortality models
References
Estimating the dimension of a model
Distribution of the Estimators for Autoregressive Time Series with a Unit Root
Specification Tests in Econometrics
Statistical analysis of cointegration vectors
Related Papers (5)
Co-integration and Error Correction: Representation, Estimation and Testing
Autoregressive conditional heteroscedasticity with estimates of the variance of United Kingdom inflation
Frequently Asked Questions (12)
Q2. What are the future works in this paper?
This paper details how the subject has advanced to its present stage of success and should convey the promise of these developments for future empirical research.
Q3. What are some of the approaches that are still extant?
A variety of approaches are still extant, such as “classical” and “Bayesian”, and with varying degrees of emphasis on least squares, instrumental variables, maximum likelihood, method of moments, and so on.
Q4. What is the main argument for establishing the link between DGPs and empirical models?
By establishing the link between DGPs and empirical models, the theory of reduction justifies both evaluative and constructive aspects of econometric modeling.
Q5. What are the implications of Anderson’s results for general-to-specific modeling?
While the construction of econometric models requires solving more complicated problems than the order of a polynomial, Anderson’s results have three direct implications for general-to-specific modeling.
Q6. What is the role of computers in the development of econometrics?
Computers have long played an essential role in the operational development of econometrics, particularly for estimation, inference, and simulation.
Q7. What are the conditions for determining the parameters of the subsystem of non-exogenous?
Non-exogenous variables are called predetermined if consistent and asymptotically unbiased estimates of the parameters of the subsystem of non-exogenous non-predetermined variables can be obtained separately from those of the subsystem of predetermined variables.
Q8. How does Phillips show that errors correlated with explanatory variables can be avoided?
Phillips shows that errors correlated with explanatory variables can be avoided by including in the estimatedequation all factors affecting directly or indirectly the explained variable as well as exogenous factors.
Q9. How many variables are irrelevant in the general unrestricted model?
To illustrate the use of a pre-search filter, suppose that the general unrestricted model has 40 variables, of which 30 are irrelevant and 10 are relevant, and where the latter have population t-values equal to two in absolute value.
Q10. What is the extent to which the model captures the salient characteristics of its corresponding local?
The extent to which the model does capture those characteristics depends both on its specification at least embedding the local DGP and on the goodness of its selection.
Q11. Why is the interpretation of test statistics at issue?
The interpretation of test statistics in the final model may be at issue because of the potential consequences of empirical model selection.
Q12. What is the procedure for comparing dynamic structural econometric models?
Hendry and Mizon’s procedure consists of formulating a congruent VAR that embeds all the contending dynamic structural econometric models (SEMs) being evaluated, and testing which of those models is a valid reduction of that VAR.