A Specification Search Algorithm for Cointegrated Systems
AbstractA specification search algorithm is proposed that aims to assist the user in the process of constructing Vector Error Correction Models. The algorithm automates testing the cointegration rank of the system and performs simplifications based on possible weak exogeneity of some variables. Overidentification restrictions can be provided by the user or can be imposed by an expert system that is a part of the algorithm. That expert system is not general, but is specific to the analysed problem. As an illustration we present an expert system for money demand specifications - it attempts to impose various restrictions typically used in money demand studies. Combinations of individual restrictions are jointly tested by a procedure similar to Omtzigt's algorithm. The algorithm is evaluated in a series of Monte-Carlo experiments. Preliminary results show that algorithm works quite well, and the true data generation processes are often recovered by the algorithm. As an empirical illustration, an automated construction of money demand systems for Germany and Switzerland is presented. Results are compared with some earlier work on money demand for these countries. In both cases stable money demand functions are found, however specifications differ from those obtained by the original authors
Download InfoTo our knowledge, this item is not available for download. To find whether it is available, there are three options:
1. Check below under "Related research" whether another version of this item is available online.
2. Check on the provider's web page whether it is in fact available.
3. Perform a search for a similarly titled item that would be available.
Bibliographic InfoPaper provided by Society for Computational Economics in its series Computing in Economics and Finance 2004 with number 321.
Date of creation: 11 Aug 2004
Date of revision:
model selection; cointegration;
Find related papers by JEL classification:
- C32 - Mathematical and Quantitative Methods - - Multiple or Simultaneous Equation Models; Multiple Variables - - - Time-Series Models; Dynamic Quantile Regressions; Dynamic Treatment Effect Models; Diffusion Processes
- C51 - Mathematical and Quantitative Methods - - Econometric Modeling - - - Model Construction and Estimation
- C52 - Mathematical and Quantitative Methods - - Econometric Modeling - - - Model Evaluation, Validation, and Selection
You can help add them by filling out this form.
CitEc Project, subscribe to its RSS feed for this item.
- David F. Hendry & Hans-Martin Krolzig, 2004.
"We Ran One Regression,"
Oxford Bulletin of Economics and Statistics,
Department of Economics, University of Oxford, vol. 66(5), pages 799-810, December.
- David F. Hendry & Hans-Martin Krolzig, 2004. "We Ran One Regression," Economics Papers 2004-W17, Economics Group, Nuffield College, University of Oxford.
- David Hendry & Hans-Martin Krolzig, 2004. "We Ran One Regression," Economics Series Working Papers 2004-W17, University of Oxford, Department of Economics.
- Jennifer Castle & David Hendry & Jurgen A. Doornik, 2010.
"Evaluating Automatic Model Selection,"
Economics Series Working Papers
474, University of Oxford, Department of Economics.
- Castle, Jennifer L. & Doornik, Jurgen A. & Hendry, David F., 2012.
"Model selection when there are multiple breaks,"
Journal of Econometrics,
Elsevier, vol. 169(2), pages 239-246.
For technical questions regarding this item, or to correct its authors, title, abstract, bibliographic or download information, contact: (Christopher F. Baum).
If references are entirely missing, you can add them using this form.