24: Advanced Diagnostics

Author
Affiliation

Smit, A. J.

University of the Western Cape

1 Introduction

Just as we revisited diagnostics for basic regression, we must do so again for the more complex models developed in recent chapters. This chapter focuses on the diagnostic techniques required to check the assumptions of mixed-effects models and other advanced structures.

2 Key Concepts

  • Residual Patterns as Feedback: Using patterns in model residuals to diagnose problems like non-linearity or missing variables.
  • Heteroscedasticity: Checking for non-constant variance in residuals, especially in hierarchical models.
  • Influence and Leverage: Identifying influential data points that may have an outsized effect on the model’s estimates.
  • Diagnosing Random Effects: Techniques for assessing the assumptions related to the random effects component of a mixed model.

This chapter is a placeholder. Content will be developed further.

Reuse

Citation

BibTeX citation:
@online{smit,_a._j.,
  author = {Smit, A. J.,},
  title = {24: {Advanced} {Diagnostics}},
  url = {http://tangledbank.netlify.app/BCB744/basic_stats/24-advanced-diagnostics.html},
  langid = {en}
}
For attribution, please cite this work as:
Smit, A. J. 24: Advanced Diagnostics. http://tangledbank.netlify.app/BCB744/basic_stats/24-advanced-diagnostics.html.