9Semiparametric mixture models
9.1 Why semiparametric mixture models?
Finite mixture models have been important tools for exploring complex data structures in many scientific areas. Parametric mixture models, being easy to interpret, quick to estimate, and having well-studied theoretical properties, have been popularly used ever since. However, similar to general parametric statistical inference tools, traditional mixture models are based on parametric model assumptions, such as linearity and normality, which are unrealistic or hard to satisfy in practice. Besides this, model misspecification could be disastrous in parametric mixture models and might lead to misleading results and inferences.
The need for semiparametric ...
Get Mixture Models now with the O’Reilly learning platform.
O’Reilly members experience books, live events, courses curated by job role, and more from O’Reilly and nearly 200 top publishers.
