Some extensions of a linear model for categorical variables.

Area of Science:

• Statistics
• Statistical Modeling
• Econometrics

Background:

• The Grizzle-Starmer-Koch (GSK) model is a statistical framework used in various fields.
• Existing models may have limitations in handling specific data distributions and constraints.
• Log-linear models and Poisson distributions are fundamental in statistical analysis.

Purpose of the Study:

• To extend the Grizzle-Starmer-Koch (GSK) model.
• To integrate traditional log-linear models and Poisson/conditional Poisson distributions.
• To define parameter estimators under general linear constraints.

Main Methods:

• Extension of the Grizzle-Starmer-Koch (GSK) model framework.
• Inclusion of log-linear models.
• Development of estimators for Poisson and conditional Poisson distributions under exact and stochastic linear constraints.

Main Results:

• The generalized Grizzle-Starmer-Koch (GSK) model now accommodates a broader range of data distributions.
• New estimators for model parameters have been successfully defined.
• The methodology supports analysis under both exact and stochastic linear constraints.

Conclusions:

• The extended Grizzle-Starmer-Koch (GSK) model offers increased flexibility for statistical analysis.
• The developed estimators provide robust methods for parameter estimation in complex scenarios.
• This work advances the application of statistical models in data analysis.