Simulating Normal Rectangle Probabilities and Their Derivatives: The Effects of Vectorization
An extensive literature in econometrics and in numerical analysis has considered the computationally difficult problem of evaluating the multiple integral representing the probability of a multivariate normal random vector constrained to lie in a rectangular region. A leading case of such an integral is the negative orthant probability, implied by the multinomial probit (MNP) model used in econometrics and biometrics. Classical parametric estimation of this model requires, for each trial parameter vector and each observation in a sample, evaluation of a normal orthant probability and its derivatives with respect to the mean vector and the variance-covariance matrix. Several Monte Carlo simulators have been developed to approximate the orthant probability integral and its linear and logarithmic derivatives that limit computation while possessing properties that facilitate their use in iterative calculations for statistical inference. In this paper, I discuss Gauss and FORTRAN implementations of 13 simulation algorithms, and I present results on the impact of vectorization on the relative computational performance of the simulation algorithms. I show that the 13 simulators differ greatly with respect to the degree of vectorizability: in some cases activating the CRAY-Y/MP4 vector facility achieves a speed-up factor in excess of 10 times, while in others the gains in speed are negligible. Evaluating the algorithms in terms of lowest simulation root-mean-squared-error for given computation time, I find that (1) GHK, an importance sampling recursive triangularization simulator, remains the best method for simulating probabilities irrespective of vectorization; (2) the crude Monte Carlo simulator CFS offers the greatest benefits from vectorization; and (3) the GSS algorithm, based on "Gibbs resampling," emerges as one of the preferred methods for simulating logarithmic derivatives, especially in the absence of vectorization.
|Date of creation:||Jul 1993|
|Date of revision:|
|Publication status:||Published in International Journal of Supercomputer Applications (1993), 7(3): 231-253|
|Contact details of provider:|| Postal: Yale University, Box 208281, New Haven, CT 06520-8281 USA|
Phone: (203) 432-3702
Fax: (203) 432-6167
Web page: http://cowles.yale.edu/
More information through EDIRC
|Order Information:|| Postal: Cowles Foundation, Yale University, Box 208281, New Haven, CT 06520-8281 USA|
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.:
- McFadden, Daniel, 1989.
"A Method of Simulated Moments for Estimation of Discrete Response Models without Numerical Integration,"
Econometric Society, vol. 57(5), pages 995-1026, September.
- Daniel McFadden, 1987. "A Method of Simulated Moments for Estimation of Discrete Response Models Without Numerical Integration," Working papers 464, Massachusetts Institute of Technology (MIT), Department of Economics.
- Paul A. Ruud., 1988.
"Extensions of Estimation Methods Using the EM Algorithm.,"
Economics Working Papers
8899, University of California at Berkeley.
- Ruud, Paul A., 1991. "Extensions of estimation methods using the EM algorithm," Journal of Econometrics, Elsevier, vol. 49(3), pages 305-341, September.
- Keane, Michael P, 1994. "A Computationally Practical Simulation Estimator for Panel Data," Econometrica, Econometric Society, vol. 62(1), pages 95-116, January.
- Vassilis A. Hajivassiliou & Axel Borsch-Supan, 1990.
"Smooth Unbiased Multivariate Probability Simulators for Maximum Likelihood Estimation of Limited Dependent Variable Models,"
Cowles Foundation Discussion Papers
960, Cowles Foundation for Research in Economics, Yale University.
- Borsch-Supan, Axel & Hajivassiliou, Vassilis A., 1993. "Smooth unbiased multivariate probability simulators for maximum likelihood estimation of limited dependent variable models," Journal of Econometrics, Elsevier, vol. 58(3), pages 347-368, August.
- Vassilis A. Hajivassiliou & Daniel L. McFadden, 1998.
"The Method of Simulated Scores for the Estimation of LDV Models,"
Econometric Society, vol. 66(4), pages 863-896, July.
- Vassilis A. Hajivassiliou & Daniel L. McFadden, 1993. "The Method of Simulated Scores for the Estimation of LDV Models," Working Papers _023, Yale University.
- V A Hajivassiliou & DL McFadden, 1997. "The Method of Simulated Scores for the Estimation of LDV Models," STICERD - Econometrics Paper Series 328, Suntory and Toyota International Centres for Economics and Related Disciplines, LSE.
- McFadden, Daniel & Ruud, Paul A, 1994. "Estimation by Simulation," The Review of Economics and Statistics, MIT Press, vol. 76(4), pages 591-608, November.
- John F. Geweke & Michael P. Keane & David E. Runkle, 1994.
"Alternative computational approaches to inference in the multinomial probit model,"
170, Federal Reserve Bank of Minneapolis.
- Geweke, John & Keane, Michael P & Runkle, David, 1994. "Alternative Computational Approaches to Inference in the Multinomial Probit Model," The Review of Economics and Statistics, MIT Press, vol. 76(4), pages 609-32, November.
- Hausman, Jerry A & Wise, David A, 1978.
"A Conditional Probit Model for Qualitative Choice: Discrete Decisions Recognizing Interdependence and Heterogeneous Preferences,"
Econometric Society, vol. 46(2), pages 403-26, March.
- J. A. Hausman & D. A. Wise, 1976. "A Conditional Profit Model for Qualitative Choice: Discrete Decisions Recognizing Interdependence and Heterogeneous Preferences," Working papers 173, Massachusetts Institute of Technology (MIT), Department of Economics.
- Vassilis A. Hajivassiliou, 1991. "Simulation Estimation Methods for Limited Dependent Variable Models," Cowles Foundation Discussion Papers 1007, Cowles Foundation for Research in Economics, Yale University.
- Pakes, Ariel & Pollard, David, 1989. "Simulation and the Asymptotics of Optimization Estimators," Econometrica, Econometric Society, vol. 57(5), pages 1027-57, September.
- Hendry, David F., 1984. "Monte carlo experimentation in econometrics," Handbook of Econometrics, in: Z. Griliches† & M. D. Intriligator (ed.), Handbook of Econometrics, edition 1, volume 2, chapter 16, pages 937-976 Elsevier.
- J. E. Dutt, 1976. "Numerical Aspects of Multivariate Normal Probabilities in Econometric Models," NBER Chapters, in: Annals of Economic and Social Measurement, Volume 5, number 4, pages 547-561 National Bureau of Economic Research, Inc.
- McCulloch, Robert & Rossi, Peter E., 1994. "An exact likelihood analysis of the multinomial probit model," Journal of Econometrics, Elsevier, vol. 64(1-2), pages 207-240.
- Stern, Steven, 1992. "A Method for Smoothing Simulated Moments of Discrete Probabilities in Multinomial Probit Models," Econometrica, Econometric Society, vol. 60(4), pages 943-52, July.
- Geweke, John, 1989. "Bayesian Inference in Econometric Models Using Monte Carlo Integration," Econometrica, Econometric Society, vol. 57(6), pages 1317-39, November.
When requesting a correction, please mention this item's handle: RePEc:cwl:cwldpp:1049. 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: (Matthew C. Regan)
If references are entirely missing, you can add them using this form.