This dissertation was written as part of the MSc in Data Science at the International
Hellenic University. The scope of this dissertation is to introduce the reader to Timeseries analysis using Reservoir Computing after establishing the core concepts of TimeSeries, Chaos theory and Reservoir Computing. Furthermore, a brief introduction of
Neural Networks is going to be discussed. More specifically, a more special tool of
Reservoir Computing, Echo State Networks, is going to be thoroughly explained and used
for the prediction of real, raw financial data. For that, sample data of both financial stocks
and indices will be used to validate the potential of the model. Finally, the accuracy of
the Echo State Networks proposed are compared to the random heuristic approach.
Collections
Show Collections
