Optimal Industrial Classification: An Application to the German Industrial Classification System
A widely used method in the analysis of large-scale econometric models is to replace the ``true model'' by an aggregative one in which the variables are grouped and replaced by sums or weighted averages of the variables in each group. The modes of aggregation of the independent and dependent variables may in principle be chosen optimally by minimizing a measure of mean-square forecast error in predicting the dependent variables from the independent variables by using the aggregative rather than detailed variables. However, this results in an optimization problem of a high degree of complexity. Nevertheless, many efficient optimization heuristics have been developed for these kinds of complex problems. We implement the Threshold Accepting heuristic for the problem of optimal aggregation of price indices in a model of the transmission of external (import and export) prices on internal prices, using German data. The algorithm and the resulting groupings are presented. The results suggest that the use of standard or ``official'' modes of aggregation will in general be far from being optimal.
|Date of creation:||01 Aug 2000|
|Date of revision:|
|Contact details of provider:|| Phone: 1 212 998 3820|
Fax: 1 212 995 4487
Web page: http://www.econometricsociety.org/pastmeetings.asp
More information through EDIRC
References listed on IDEAS
Please report citation or reference errors to , or , if you are the registered author of the cited work, log in to your RePEc Author Service profile, click on "citations" and make appropriate adjustments.:
- Winker, Peter, 1994.
"Identification of multivariate AR-models by threshold accepting,"
Discussion Papers, Series II
224, University of Konstanz, Collaborative Research Centre (SFB) 178 "Internationalization of the Economy".
- Winker, Peter, 1995. "Identification of multivariate AR-models by threshold accepting," Computational Statistics & Data Analysis, Elsevier, vol. 20(3), pages 295-307, September.
- Peter Winker, 2000. "Optimized Multivariate Lag Structure Selection," Computational Economics, Society for Computational Economics, vol. 16(1/2), pages 87-103, October.
- Pesaran, M Hashem & Pierse, Richard G & Kumar, Mohan S, 1989.
"Econometric Analysis of Aggregation in the Context of Linear Prediction Models,"
Econometric Society, vol. 57(4), pages 861-88, July.
- M. H. Pesaran & R. G. Pierse & M. S. Kumar, 1988. "Econometric Analysis of Aggregation in the Context of Linear Prediction Models," UCLA Economics Working Papers 485, UCLA Department of Economics.
- Geweke, John, 1985. "Macroeconometric Modeling and the Theory of the Representative Agent," American Economic Review, American Economic Association, vol. 75(2), pages 206-10, May.
- Thompson, Gary D. & Lyon, Charles C., 1992. "A generalized test for perfect aggregation," Economics Letters, Elsevier, vol. 40(4), pages 389-396, December.
- Edward E. Leamer, 1982. "Optimal Aggegation of Linear Systems," UCLA Economics Working Papers 240, UCLA Department of Economics.
- Winker, Peter, 1992. "Some notes on the computational complexity of optimal aggregation," Discussion Papers, Series II 184, University of Konstanz, Collaborative Research Centre (SFB) 178 "Internationalization of the Economy".
- Hirotugu Akaike, 1969. "Fitting autoregressive models for prediction," Annals of the Institute of Statistical Mathematics, Springer, vol. 21(1), pages 243-247, December.
- Cotterman, R & Peracchi, F, 1992. "Classification and Aggregation: An Application to Industrial Classification in CPS Data," Journal of Applied Econometrics, John Wiley & Sons, Ltd., vol. 7(1), pages 31-51, Jan.-Marc.
When requesting a correction, please mention this item's handle: RePEc:ecm:wc2000:0522. See general information about how to correct material in RePEc.
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.