Computational Economics

, Volume 25, Issue 4, pp 343–379 | Cite as

Solving Finite Mixture Models: Efficient Computation in Economics Under Serial and Parallel Execution

  • Christopher FerrallEmail author


Many economic models are completed by finding a parameter vector θ that optimizes a function f(θ), a task that can only be accomplished by iterating from a starting vector θ0. Use of a generic iterative optimizer to carry out this task can waste enormous amounts of computation when applied to a class of problems defined here as finite mixture models. The finite mixture class is large and important in economics and eliminating wasted computations requires only limited changes to standard code. Further, the approach described here greatly increases gains from parallel execution and opens possibilities for re-writing objective functions to make further efficiency gains.


heterogeneous agent models numerical optimization 


Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.


  1. Arcidiacono, P. and Jones, J.B. (2003). Finite mixture distributions, sequential likelihood and the EM algorithm. Econometrica, 71(3), 933–946.CrossRefGoogle Scholar
  2. Aguirregabiria, V. and Mira, P. (2002). Swapping the nested fixed point algorithm: A class of estimators for discrete Markov decision models. Econometrica, 70(4), 1519–1543.CrossRefMathSciNetGoogle Scholar
  3. Debreu, G. (1983). A social equilibrium existence theorem. In Mathematical Economics: Twenty Papers of Gerard Debreu, Econometric Society Monographs 4, Cambridge University Press.Google Scholar
  4. Doornik, J.A., Hendry, D.F. and Shephard, N. (2002). Computationally-intensive econometrics using a distributed matrix-programming language. Philosophical Transactions of the Royal Society of London, Series A, 360, 1245–1266.Google Scholar
  5. Eckstein, Z. and Wolpin, K.I. (1990). Estimating a market equilibrium search model from panel data on individuals. Econometrica, 58(4), 783–808.Google Scholar
  6. Goldfeld, S.M., Quandt, R.E. and Trotter, H.F. (1966). Maximization by quadratic hill-climbing. Econometrica, 34(3), 541–551.Google Scholar
  7. Heckman, J.J. and Singer, B. (1984). A method for minimizing the imapct of distributional assumptions in econometric models for duration data. Econometrica, 52Google Scholar
  8. Hotz, J. and Miller, R.A. (1993). Conditional choice probabilities and the estimation of dynamic models. Reviews of Economic Studies, 60, 497–529.Google Scholar
  9. Imai, S., Jain, N. and Ching, A. (2002). Bayesian estimation of dynamic discrete choice models. Manuscript, Ohio State, April,
  10. Judd, K.L. (1988). Numerical Methods In Economics. MIT Press, Cambridge.Google Scholar
  11. Lee, D. (2005). An estimable dynamic general equilibrium model of work, schooling and occupational choice. International Economic Review, 46, 1–34.CrossRefMathSciNetGoogle Scholar
  12. Mcfadden, D. and Train, K. (2000). Mixed MNL models for discrete response. Journal of Applied Econometrics, 15, 447–470.CrossRefGoogle Scholar
  13. Nagurney, A. (1996). Parallel computation. In D. Kendrick, J. Rust and H.M. Amman (eds.), Handbook of Computational Economics, 331–400. North-Holland, Amsterdam.Google Scholar
  14. Press, W.H., Flannery, B.P., Teukolsky, S.A. and Vetterling, W.T. (1987). Numerical Recipes: The Art of Scientific Computing. Cambridge University Press, New York.Google Scholar
  15. Rios-Rull, V. (1999). Computation of equilibria in heterogenous agent models. In R. Marimon and A. Scott (eds.), Computaional Methods for the Study of Dynamic Economies: An Introduction. Oxford University Press, Oxford.Google Scholar
  16. Rust, J. (1994). Structure estimation of markov decision processes. In R. Engle and D. McFadden (eds.), Handbook of Econometrics, Vol. 4, 30823139. North Holland, Amsterdam.Google Scholar
  17. Rust, J. (1996). Numerical dyanmic programming in economics. In H. Amman, D. Kendric and J. Rust (eds.), Handbook of Computational Economics. Elsevier, North Holland.Google Scholar
  18. Schittkowski, K. (2001). NLPQLP: A new Fortran implementation of a sequential quadratic programming algorithm for parallel computing. Manuscript,
  19. Swann, C.A. (2001). Software for parallel computing: The LAM implementation of MPI. Journal of Applied Econometrics, 16(2), 185–194.CrossRefGoogle Scholar
  20. Swann, C.A. (2002). Maximum likelihood estimation using parallel computing: An intorduction to MPI. Computational Economics, 19(2), 145–178.CrossRefGoogle Scholar

Copyright information

© Springer Science + Business Media, Inc. 2005

Authors and Affiliations

  1. 1.Department of Economics, Dunning HallQueen’s UniversityKingstonCanada

Personalised recommendations