Providing software solutions since 1976
Capabilities
- Analysis of Variance
- Bayesian Analysis
- Categorical Data Analysis
- Cluster Analysis
- Descriptive Statistics
- Distribution Analysis
- Discriminant Analysis
- Exact Methods
- Group Sequential Design and Analysis
- Market Research
- Missing Value Imputation
- Mixed Models
- Multivariate Analysis
- Nonparametric Statistics
- Power and Sample Size
- Psychometric Analysis
- Regression
- Spatial Analysis
- Structural Equations
- Survey Sampling and Analysis
- Survival Analysis
- Transformations
- SAS/STAT Procedures A-Z
The FMM Procedure (Experimental)
The FMM procedure fits statistical models to data for which the distribution of the response is a finite mixture of univariate distributions–that is, each response comes from one of several random univariate distributions with unknown probabilities.
The following are highlights of the FMM procedure's capabilities:
- model the component distributions in addition to the mixing probabilities
- fits finite mixture models by maximum likelihood or Bayesian methods
- fits finite mixtures of regression and generalized linear models
- enables you to define the model effects for the mixing probabilities and their link function
- enables you to model overdispersed data
- enables you to estimate multimodal or heavy-tailed densities
- fits zero-inflated or hurdle models to count data with excess zeros
- fits regression models with complex error distributions
- classifies observations based on predicted component probabilities
- supports twenty different response distributions
- supports linear equality and inequality constraints on model parameters
- enables you to specify the response variable by using either the response syntax or the events/trials syntax
- supports automated model selection for homogeneous mixtures
- supports weighted estimation
- enables you to control the performance characteristics of the procedure (for example, the number of CPUs, the number of threads for multithreading, and so on)
- enables you to obtain separate analyses on observations in groups
- enables you to create a data set that contains observationwise statistics that are computed after fitting the model
- uses ODS to create a SAS data set corresponding to any table
- supports ODS Graphics
For further details see the SAS/STAT User's Guide:
The FMM Procedure
( PDF | HTML )
Examples
- Mixture Modeling for Binomial Overdispersion: "Student," Pearson, Beer, and Yeast
- Modeling Zero-Inflation: Is it Better to Fish Poorly or Not to Have Fished At All?
- Looking for Multiple Modes: Are Galaxies Clustered?
- Modeling Mixing Probabilities: All Mice Are Created Equal, but Some Are More Equal
- The Usefulness of Custom Starting Values: When Do Cows Eat?
- Enforcing Homogeneity Constraints: Count and Dispersion–It Is All Over!
Statistics and Operations Research Home Page | SAS/STAT Software
