An Adaptive Retraining Method for the Exchange Rate Forecasting

The paper advances an original artificial intelligence-based mechanism for specific economic predictions. The time series under discussion are non-stationary; therefore the distribution of the time series changes over time. The algorithm establishes how a viable structure of an artificial neural network (ANN) at a previous moment of time could be retrained in an efficient manner, in order to support modifications in a complex input-output function of financial forecasting. A "remembering process" for the former knowledge achieved in the previous learning phase is used to enhance the accuracy of the predictions.
The results show that the first training (which includes the searching phase for the optimal architecture) always takes a relatively long time, but then the system can be very easily retrained, as there are no changes in the structure. The advantage of the retraining procedure is that some relevant aspects are preserved (remembered) not only from the immediate previous training phase, but also from the previous but one phase, and so on. A kind of slow forgetting process also occurs; thus it is much easier for the ANN to remember specific aspects of the previous training instead of the first training.
The experiments reveal the high importance of the retraining phase as an upgrading/updating process and the effect of ignoring it, as well. There has been a decrease in the test error when successive retraining phases were performed. 

Keywords: Neural Networks, Exchange Rate, Adaptive Retraining, Delay Vectors, Iterative Simulation
JEL Classification: C45, C53, F47