Mixture Models in Statistics

Research output: Chapter in Book/Report/Conference proceedingEntry for encyclopedia/dictionarypeer-review


Mixture models are a discrete or continuous weighted combination of distributions aimed at representing a heterogeneous population comprising of several (two or more) distinct subpopulations. Such models are often used to approximate nonstandard distributions when the underlying population can be regarded as homogenous. A mathematical definition of mixture models is provided, several of the diverse applications of such models are outlined, and the main methods of estimating the model's parameters are discussed.
Original languageEnglish
Title of host publicationInternational Encyclopedia of the Social & Behavioral Sciences
PublisherElsevier Science Ltd.
ISBN (Print)9780080430768
StatePublished - 2001


Dive into the research topics of 'Mixture Models in Statistics'. Together they form a unique fingerprint.

Cite this