DOCUMENT RESUME ED 359 262 TM 020 031 AUTHOR Wang, Yuh-Yin Wu; Schafer, William D. TITLE Maximum Likelihood and Minimum Distance Applied to Univariate Mixture Distributions. PUB DATE Apr 93 NOTE 45p.; Paper presented at the Annual Meeting of the American Educational Research Association (Atlanta, GA, April 12-16, 1993). PUB TYPE Reports Evaluative/Feasibility (142) Speeches /Conference Papers (150) EDRS PRICE MF01/PCO2 Plus Postage. DESCRIPTORS Comparative Analysis; *Computer Simulation; Equations (Mathematics); *Estimation (Mathematics); Graphs; *Mathematical Models; *Maximum Likelihood Statistics; Monte Carlo Methods; Statisti:-.al Distributions IDENTIFIERS EM Algorithm; *Minimum Distance Principle; Mixtures; *Univariate Analysis ABSTRACT This Monte-Carlo study compared modified Newton (NW), expectation-maximization algorithm (EM), and minimum Cramer-von Mises distance (MD), used to estimate parameters of univariate mixtures of two components. Data sets were fixed at size 160 and manipulated by mean separation, variance ratio, component proportion, and non-normality. Results indicate that NW is the poorer estimation procedure. Eh is less sensitive to different initial inputs and produced the lowest singularity rate. MD is more robust to non-normality and to incorrect model assumption of variance. In practice, MD is recommended. The singularity problem is not severe enough to be a practical concern. (Twelve figures present details of the simulations and analyses.) (Author/SLD) *********************************************************************** Reproductions supplied by EDRS are the best that can be made from the original document. ***********************************************************************