Information-theoretic methods work in a series of steps:
1. A set of candidate models are selected that represent different hypotheses for explaining reality.
3. The data are used to assess the relative support for the different models, by estimating the amount of information lost when using each. The best model is selected as the one that is estimated to lose the least amount of information.
4. Any required predictions are made using an average of the models that is weighted towards those that are estimated to lose less information.
Was this article helpful?