Type-2 Fuzzy Clustering and a Type-2 Fuzzy Inference Neural Network for the Prediction of Short-term Interest Rates

David Enke, Nijat Mehdiyev

In: Procedia Computer Science 20 Pages 115-120 Elsevier 2013.


The following paper discusses the use of a hybrid model for the prediction of short-term US interest rates. The model consists of a differential evolution-based fuzzy type-2 clustering with a fuzzy type-2 inference neural network, after input preprocessing with multiple regression analysis. The model was applied to forecast the US 3- Month T-bill rates. Promising model performance was obtained as measured using root mean square error.

Weitere Links

German Research Center for Artificial Intelligence
Deutsches Forschungszentrum für Künstliche Intelligenz