Efficiency of the Cyber 205 for Stochastic Simulations of a Simultaneous,Nonlinear,Dynamic Econometric Model

Abstract

A code for carrying out stochastic simulations of the MPS econometric model, a simultaneous, nonlinear, dynamic system of equations involving approximately 600 variables, was adapted to run on the CYBER 205 vector processor. The effi ciency gained relative to performance on an IBM 3081-GX was significant: the CPU time re quired for a given simulation was reduced by a factor of 10 to 12. However, this is considerably smaller than the theoretical gain expected from the manufacturer's ratings. The basic causes of this smaller gain are the fairly small real memory of the CYBER 205 and the inefficiency of the CYBER 205 vector processor in performing ele mentary function look-up. The primitive nature of the operating system and some aspects of the FORTRAN compiler on the CYBER 205 required more reprogramming of our large code than such a conversion should demand. However, potential gains from such a conversion will be great once the deficiencies described above are eliminated.

Get full access to this article

View all access options for this article.

References

Agarwal, R.C. , Kooley, J.W. , Gustavson, F.G. , et al. 1986. New scalar and vector elementary functions for the IBM system 370. IBM J. Res. Develop. 30(2):126-143.

Brayton, F. , and Mauskopf, E. 1985. The Federal Reserve Board MPS quarterly econometric model of the U.S. economy. Econom. Modelling 2(3):170-292.

Don, F.J.H. 1986. Linear methods in non-linear models. Rotterdam, The Netherlands: Centraal Planbureau.

Efron, B. 1979. Bootstrap methods: another look at the jacknife . Ann. Statist. 7:1-26.

Fair, R. 1984. Specification, estimation, and analysis of macroeconometric models. Cambridge: Harvard University Press.

Gabay, D. , Nepomiastchy, P. , Rachdi, M. , and Ravelli, A. 1980. Numerical methods for simulation and optimal control of large scale macroeconomic models. In: Applied stochastic control in econometrics and management science, edited by A. Bensoussan , P. Kleindorfer , and C. S. Tapiero. Amsterdam : North-Holland.

Gallion, S. , Levy, R.M. , Weiner, P.K. , et al. 1986. Implementation of a macromolecular mechanics program on a CYBER 205 supercomputer. Comput. and Chemistry 10(3):165-173.

Gill, P. , Murray, W. , and Wright, M. 1981. Practical optimization. New York: Academic Press.

Henaff, P. 1986. A Newton's algorithm for solving multicountry econometric models. Cambridge: Massachusetts Institute of Technology, CCREMS.

10.

Henaff, P. 1985. Numerical comparison of some Newton-like methods . Austin: University of Texas, Department of Economics, doctoral dissertation.

11.

Henaff, P. , and Norman, A.L. 1986. Solving nonlinear econometric models using vector processors . In: Computational science and statistics: proceedings of the 18th symposium on the interphase, edited by T. A. Boardman. Washington, D.C.: American Statistical Association.

12.

Nepomiastchy, P. , and Ravelli, A. 1978. Adapted methods for solving and optimizing quasi-triangular econometric models. Ann. Econom. Social Measurement 6:555-562.

13.

Rice, J.R. 1983. Numerical methods, software and analysis: IMSL reference edition. New York: McGraw-Hill .

14.

Tinsley, P. , and Fries, G. 1981. Two methods of generating pseudo-historical forecast errors . Washington, D.C.: Board of Governors of the Federal Reserve System.