Model selection: from the Null Hypothesis Testing to the AIC (Akaike Information Criterion)
Are you already subscribed?
Login to check
whether this content is already included on your personal or institutional subscription.
Abstract
Models are widely used in the psychological research whereas the weakness of the null hypothesis testing in the model selection has been pointed out. The Information Theory approach has started a new process of selecting models, and it has contributed to the definition of alternative methods such as the AIC (Akaike Information Criterion). The present paper aimed to describe the characteristics of the AIC and to exemplify how it is simple to calculate and interpret the AIC and its transformations.
Keywords
- Model selection
- Hypothesis testing
- AIC
- Akaike Weights
- Information theory