The Cluster-Weighted Model (CWM) is a member of the family of the Mixtures of Regression Models and it is referred as Mixture of Regression with Ran- dom Covariates. Currently, the only procedure for estimating these models is R pack- age flexcwm. The aim of this article is to introduce a new software component, the Stata package cwmglm which estimates CWMs based on the most common general- ized linear models. Our software also extends to Stata users the possibility of estimat- ing parsimonious models of Gaussian distributions with alternative specifications of the variance matrix. cwmglm also calculates the the generalized coefficients of de- termination and bootstrap standard errors that are not currently available in flexcwm. We illustrate the use of cwmglm with real data on Covid-19 admissions.

A Stata implementation of Cluster Weighted Models: the CWMGLM Package

Salvatore Ingrassia
Methodology
;
Giorgio Vittadini
Methodology
2023-01-01

Abstract

The Cluster-Weighted Model (CWM) is a member of the family of the Mixtures of Regression Models and it is referred as Mixture of Regression with Ran- dom Covariates. Currently, the only procedure for estimating these models is R pack- age flexcwm. The aim of this article is to introduce a new software component, the Stata package cwmglm which estimates CWMs based on the most common general- ized linear models. Our software also extends to Stata users the possibility of estimat- ing parsimonious models of Gaussian distributions with alternative specifications of the variance matrix. cwmglm also calculates the the generalized coefficients of de- termination and bootstrap standard errors that are not currently available in flexcwm. We illustrate the use of cwmglm with real data on Covid-19 admissions.
2023
9788891935632
cluster weighted models
clustering
parsimonious models
Stata
File in questo prodotto:
Non ci sono file associati a questo prodotto.

I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.

Utilizza questo identificativo per citare o creare un link a questo documento: https://hdl.handle.net/20.500.11769/618450
Citazioni
  • ???jsp.display-item.citation.pmc??? ND
  • Scopus ND
  • ???jsp.display-item.citation.isi??? ND
social impact