A new model selection procedure for finite mixture regression models

2019 ◽  
Vol 49 (18) ◽  
pp. 4347-4366 ◽  
Author(s):  
Conglian Yu ◽  
Xiyang Wang
Keyword(s):  
Model Selection ◽  
Regression Models ◽  
Selection Procedure ◽  
Finite Mixture ◽  
New Model ◽  
Mixture Regression ◽  
Model Selection Procedure ◽  
Finite Mixture Regression ◽  
Mixture Regression Models

scholarly journals Modelling semi-continuous data using mixture regression models with an application to cattle production yields

2011 ◽  
Vol 150 (1) ◽  
pp. 109-121 ◽  
Author(s):  
E. J. BELASCO ◽  
S. K. GHOSH
Keyword(s):  
Regression Models ◽  
Continuous Variable ◽  
Continuous Variables ◽  
Multivariate Models ◽  
Mixture Regression ◽  
Potential Applications ◽  
Mixture Regression Models ◽  
Mixture Regression Model ◽  
The Relationship ◽  
Positive Half

SUMMARYThe present paper develops a mixture regression model that allows for distributional flexibility in modelling the likelihood of a semi-continuous outcome that takes on zero value with positive probability while continuous on the positive half of the real line. A multivariate extension is also developed that builds on past multivariate models by systematically capturing the relationship between continuous and semi-continuous variables, while allowing for the semi-continuous variable to be characterized by a mixture model. The flexibility associated with this model provides potential applications in many production system studies. The empirical model is shown to provide a more accurate measure of mortality rates in cattle feedlots, both independently and within a system including other performance and health factors.

Sign in / Sign up

Export Citation Format

Share Document