Recurrent SOM with local linear models in time series prediction

Recurrent Self-Organizing Map (RSOM) is studied in three di erent time series prediction cases. RSOM is used to cluster the series into local data sets, for which corresponding local linear models are estimated. RSOM includes recurrent di erence vector in each unit which allows storing context from the past input vectors. Multilayer perceptron (MLP) network and autoregressive (AR) model are used to compare the prediction results. In studied cases RSOM shows promising results.

• Type: Conference
• Year: 1998
• Where: ESANN
• Authors: Timo Koskela, Markus Varsta, Jukka Heikkonen, Kimmo Kaski
• Keywords: ESANN 1998, ESANN 2007, Recurrent Di Erence, Recurrent Self-Organizing Map, Series Prediction Cases