Can I read Multiple quantitative predictors, dealing with large models, and Bayesian ANOVA on EtoBox?

Multiple quantitative predictors, dealing with large models, and Bayesian ANOVA by Santiago Barreda; Noah Silbert is a book available to read on EtoBox.

What is Multiple quantitative predictors, dealing with large models, and Bayesian ANOVA about?

In this chapter, we introduce models with multiple quantitative predictors and interactions between them. After that, we will have covered the basic modeling concepts necessary to fit most linear models. The models you fit to your own data will include some combination of the elements covered in the previous chapters, and can potentially result in large models with hundreds (or thousands) of estimated parameters. Traditionally, models with many estimated parameters have had three general problems:1 A model with many predictors may have 'too many' predictors. We will vaguely define 'extra' predictors as those which have no meaningful statistical association with your dependent variable, and which do not appreciably improve your model in any way. Sometimes, the 'extra' predictors have estimated values that are difficult to distinguish from those of the 'real' parameters, leading to incorrect conclusions. 2 A model with many parameters may have more difficulties with fitting/convergence, meaning you can't get good estimates for the model coefficients. Regardless of the approach to parameter estimation, more complicated models make it more difficult to find the optimal parameter values

Author

Santiago Barreda; Noah Silbert

Publisher

Routledge

Published

2023

Language

EN

ISBN

9781032259628

Subjects

Psychology, Science, Mathematics