Communications for Statistical Applications and Methods

The Korean Statistical Society (한국통계학회)
권호 표지

A Penalized Likelihood Method for Model Complexity

  • Ahn, Sung M. (School of Management Information Systems, Kookmin University)
  • Published : 2001.04.01
Download PDF
⟨ Previous Next ⟩

Abstract

We present an algorithm for the complexity reduction of a general Gaussian mixture model by using a penalized likelihood method. One of our important assumptions is that we begin with an overfitted model in terms of the number of components. So our main goal is to eliminate redundant components in the overfitted model. As shown in the section of simulation results, the algorithm works well with the selected densities.

Keywords

References

  1. Neural Networks for Pattern Recognition Bishop,C.M.
  2. Journal of Royal Statistical Society(B) v.39 Maximum Likelihood from Incomplete Data via the EM Algorithm Dempster,A.P.;Laird,N.M.;Rubin,D.B.
  3. Annals of Statistics v.20 no.2 Exact Mean Integrated Squared Error Marron,J.S.;Wand,M.P.
  4. IEEE Transactions on Neural Networks v.9 no.4 Averaging, Maximum Penalized Likelihood and Bayesian Estimation for Improving Gaussian Mixture Probability Density Estimates Ormoneit,D.;Tresp,V.
  5. Proceedings of the IEEE v.78 no.9 Networks for approximation and Learning Poggio,T.;Girosi,F.
  6. Journal of American Statistical Association v.89 no.427 Adaptive Mixtures Priebe,C.E.
  7. Journal of American Statistical Association v.92 Practical Bayesian Density Estimation Using Mixtures of Normals Roeder,K.;Wasserman,L.
  8. Journal of Computational and Graphical Statistics v.4 A Visualization Technique for Studying the Iterative Estimation of Mixture Densities Solka,J.L.;Poston,W.L.;Wegman,E.J.
  9. Statistical Analysis of Finite Mixture Distributions Titterington,D.M.;Smith,A.F.M.;Makov,U.E.
  10. Neural Computation v.8 On Convergence Properties of the EM Algorithm for Gaussian Mixtures Xu,L.;Jordan,M.I.