Finite mixture densities can be used to model data from populations known or suspected to contain a number of separate subpopulations. Most commonly used are mixture densities with Gaussian (univariate or multivariate) components, but mixtures with other types of component are also increas ingly used to model, for example, survival times. This paper gives a general introduction to the topic which should help when considering the other more specialized papers in this issue.
McLachlan GJ, Basford KEMixture models; inference and applications to clustering. New York: Marcel Dekker, Inc., 1988 .
3.
McGiffin DC, Galbraith AJ, McLachlan GJ et al. Aortic valve infection-risk factors for death and recurrent endocarditis following aortic valve replacement. Journal of Thoracic Cardiovascular Surgery1992; 104: 511-20.
4.
Blackstone EH , Naftel DC, Turner METhe decomposition of time-varying hazard into phases, each incorporating a separate stream of concomitant information. Journal of the American Statistical Association1986; 81: 615-24.
5.
Pickering RM, Forbes JFA classification of Scottish infants using latent class analysis. Statistics in Medicine1984; 3: 249-59.
6.
Pearson K.Contribution to the mathematical theory of evolution. Philosophical Transactions A1894; 185: 71-110.
7.
Tan WY, Chang WCSome comparisons of the method of moments and the method of maximum likelihood in estimating parameters of a mixture of two normal densities. Journal of the American Statistical Association1972; 67: 702-708.
8.
Charlier Cvl.Researchers into the theory of probability. Lunds Universities Årskrift, Ny foljd1906; 2.1, No. 5.
9.
Strömgren B.Tables and diagrams for dissecting a frequency curve into components by the half-invariant method. Skand. Aktuarietidskr1934; 17: 7-54.
10.
Rao CRThe utilization of multiple measurements in the problems of biological classification . Journal of the Royal Statistical Society, Series B 1948; 10: 159-203.
11.
Charlier Cvl , Wicksell SDOn the dissection of frequency functions. Arkiv. for Matematik, Astronomi och Fysik1924; 18: no. 6.
12.
Doetsch G.Die Elimination des Dopplereffekts bei spektroskopischen feinstrukturen and exakte Bestimming der Kompronenten. Zeitschrift für Physik1928; 49: 705-30.
13.
Tiselius A. , Kabat EAElectrophoretic study of immune sera and purified antibody preparations. Journal of Experimental Medicine1939; 69: 119-31.
14.
Harding JPThe use of probability paper for the graphical analysis of polymodal frequency distributions. Journal of the Marine Biology Association of the UK1949; 28: 141-53.
15.
Hald A.Statistical theory with engineering applications. New York : Wiley, 1952.
16.
Cassie RMSome uses of probability paper in the analysis of size frequency distributions . Australian Journal of Marine and Freshwater Research1954; 5: 513-22.
17.
Bhattacharya CGA simple method of resolution of a distribution into Gaussian components. Biometrics1967; 23: 115-35.
18.
Newcombe S.A generalized theory of the combination of observations so as to obtain the best result. American Journal of Mathematics1886; 8: 343-66.
19.
Hasselblad V.Estimation of parameters for a mixture of normal distributions. Technometrics1966; 8: 431-44.
20.
Hasselblad V.Estimation of finite mixtures of distribution from the exponential family. Journal of the American Statistical Association1969; 64: 1459-71.
21.
Wolfe JHA computer program for the maximum likelihood analysis of types. Technical Bulletin, 65-15: San Diego: US Naval Personnel Research Activity, 1965.
22.
Wolfe JHNORMIX: computational methods for estimating the parameters of multivariate normal mixtures of distributions. Research Memorandum, SRM 68-2. San Diego: US Naval Personnel Research Activity, 1967.
Dempster AP , Laird NM, Rubin DBMaximum likelihood from incomplete data via the EM algorithm. Journal of the Royal Statistical Society, Series B1977; 39: 1-38.
25.
Medgyessi P.Decompositions of superpositions of distribution functions. Budapest: Hungarian Academy of Sciences, 1961.
26.
Blischke WRMixtures of distributions, In: International encyclopedia of statistics, Kruskal WH, Tanur JM eds. New York: The Free Press, 1979.
27.
Thomas Eac.Mathematical models for the clustered firing of single cortical neurones. British Journal of Mathematical and Statistical Psychology1966; 19: 151-62.
28.
Joffe ADMixed exponential estimation by the method of half moments. Applied Statistics1964; 13: 91-98.
29.
Mendenhall W., Hader RJEstimation of parameters of mixed exponentially distributed failure time distributions from censored life test data. Biometrika1958; 45: 504-20.
30.
Kao Jhk.A graphical estimation of mixed Weibull parameters in life-testing electron tubes. Technometrics1959; 1: 389-407.
31.
Lui KJ, Darrow WW, Rutherford GWA model-based estimate of the mean incubation period for AIDS in homosexual men. Science1988; 20: 1333-35.
32.
McLachlan GJ , McGiffin DC. On the role of finite mixture models in survival analysis. Statistical Methods in Medical Research1994; 2: 211-26.
33.
Green BFA general solution for the latent class model of latent structure analysis . Psychometrika1951; 16: 151-66.
Lazarsfeld PF, Henry NWLatent structure analysis. New York: Houghton Mifflin, 1968.
36.
Aitkin M., Anderson D., Hinde J.Statistical modelling of data on teaching styles. Journal of the Royal Statistical Society, Series A 1981; 144: 419-48.
37.
Cohen ACEstimation in mixture of two normal distributions. Technometrics1967; 9: 15-28.
38.
Everitt BS, Hand DJFinite mixture distributions. London: Chapman and Hall, 1981.
39.
Quandt RE, Ramsey JBEstimating mixtures of normal distributions and switching regressions. Journal of the American Statistical Association1978; 73: 730-38.
40.
Lindsay BG, Basak P.Multivariate normal mixtures: a fast consistent method of moments. Journal of the American Statistical Association1993; 88: 468-76.
41.
Day NEEstimating the components of a mixture of normal distributions. Biometrika1969; 56: 463-74.
42.
Jones PN, McLachlan GJ. Improving the convergence rate of the EM algorithm for a mixture model fitted to grouped truncated data. Journal of Statistical Computation and Simulation1992; 43: 31-44.
43.
Keifer J., Wolfowitz J.Consistency of the maximum likelihood estimates in the presence of infinitely many incidental parameters. Annals of Mathematical Statistics1956; 27: 887-906.
44.
Hathaway RJA constrained formulation of maximum-likelihood estimation for normal mixture distributions. Annals of Statistics1985; 13: 795-800.
45.
Everitt BSCluster analysis. London: Arnold , 1993.
Behboodian J.Information matrix for a mixture of two normal distributions. Journal of Statistical Computation andSimulation1972; 1: 295-314.
48.
Chang WCConfidence interval estimation and transformation of data in a mixture of two multivariate normal distributions with any given large dimension. Technometrics1979; 21: 351-55.
49.
Holgersson N., Jorner U.Decomposition of a mixture of two normal components . Research Report76-13, University of Uppsala, Sweden, 1976.
50.
Fowlkes EBSome methods for studying the mixture of two normal (lognormal) distributions . Journal of the American Statistical Association1979; 74: 561-75.
51.
Wolfe JHA Monte Carlo study of the sampling distribution of the likelihood ratio for mixtures of multinormal distributions. Technical Bulletin STB72-2. San Diego: Naval Personnel and Training Research Laboratory , 1971.
52.
Hernandez Avila A.Problems in cluster analysis. [Thesis]. Oxford, 1979.
53.
Everitt BSA Monte Carlo investigation of the likelihood ratio test for the number of components in a mixture of normal distributions. Multivariate Behavioural Research1981; 16: 171-80.
54.
Thode HC, Finch SJ, Mendell NRSimulated percentage points for the null distribution of the likelihood ratio test for a mixture of two normals. Biometrics1989; 44: 1195-1201.
55.
McLachlan GJOn bootstrapping the likelihood ratio test statistic for the number of components in a normal mixture. Applied Statistics1987; 36: 318-24.
56.
Gosh JM, Sen PKOn the asymptotic performance of the log-likelihood ratio statistic for the mixture model and related results. In: Proceedings of the Berkeley Conference in Honour of Jerzy Neyman and Jack Keifer, Volume II, Le Cam LM , Olshen RA eds. Monterey: Wadsworth, 1985.
57.
Self SG, Liang KYAsymptotic properties of maximum likelihood estimators and likelihood ratio tests under nonstandard conditions. Journal of the American Statistical Association1987; 82: 605-10.
58.
Kraeplin E.Dementia praecox and paraphenia. Edinburgh: Livingstone, 1919.
59.
Lewine Rrj.Sex differences in schizophrenia: timing or subtypes? PsychologicalBulletin1981; 90: 432-44.
60.
Betemps EJ, Buncher CRBirthplace as a risk factor in motor neurone disease and Parkinson's disease . International Journal of Epidemiology1993 ; 22: 898-904.
61.
Fergusson DM , Horwood LJA latent class model of smoking experimentation in children. Journal of Child Psychology and Psychiatry1989; 30: 761-73.
62.
Everitt BSA finite model for the clustering of mixed mode data. Statistics and Probability Letters1988; 6: 305-309.
63.
Lawrence CJ, Krzanowski WJMixture separation for mixed-mode data. Statistics and Computing1995, 6: 85-92.
64.
Titterington DMMixture distributions (update). In: Kotz SM ed. Encyclopedia of statistical science (update). New York: Wiley, 1996.
65.
Ripley BDNeural networks and related models for classification (with discussion). Journal of the Royal Statistical Society, Series B1994; 56: 409-56.
66.
Jorgensen MAInfluence-based diagnostics for finite mixture models. Biometrics1990; 46: 1047-58.
67.
Lindsay BG, Roeder K.Residual diagnostics for mixture models. Journal of the American Statistical Association1992; 87: 785-94.
68.
Roeder K.A graphical technique for determining the number of components in a mixture of normals. Journal of the American Statistical Association1994; 89: 487-95.
