CLME: An R Package for Linear Mixed Effects Models under Inequality Constraints

Area of Science:

• Statistics
• Computational Statistics
• Econometrics

Background:

• Statistical inference for inequality constraints is crucial in linear fixed and mixed effects models.
• Existing methods lack user-friendly software implementations, particularly for these model types.
• Inequality constraints are common in various research applications.

Purpose of the Study:

• Introduce CLME, an R package for testing diverse inequality constraints.
• Provide a user-friendly tool for statistical inference in linear mixed models.
• Address the lack of accessible software for inequality constraint testing.

Main Methods:

• Utilizes a residual bootstrap methodology for robust inference.
• The CLME package in R implements a broad collection of inequality constraint tests.
• The method is robust to non-normality and heteroscedasticity.

Main Results:

• Demonstrates the utility of CLME through two illustrative datasets.
• Provides a practical R package for inequality constraint testing.
• The residual bootstrap approach ensures reasonable robustness.

Conclusions:

• CLME offers a valuable and accessible tool for researchers.
• Facilitates statistical inference under inequality constraints in mixed models.
• The package includes a graphical interface built with the shiny package for enhanced usability.