Online Learning with Radial Basis Function Networks
This addresses forecasting challenges in financial time series, but it appears incremental as it builds on existing methods like radial basis function networks and online learning.
The paper tackled forecasting financial time series with nonstationarity and autocorrelation by combining feature representation transfer with sequential optimization for multi-horizon returns forecasts, resulting in an online learning rbfnet that outperformed a random-walk baseline and several batch learners.
Financial time series are characterised by their nonstationarity and autocorrelation. Even if these time series are differenced, technically ensuring their stationarity, they experience regular covariate shifts and concept drifts. Against this backdrop, we combine feature representation transfer with sequential optimisation to provide multi-horizon returns forecasts. Our online learning rbfnet outperforms a random-walk baseline and several powerful batch learners. The rbfnets we formulate are naturally designed to measure the similarity between test samples and continuously updated prototypes that capture the characteristics of the feature space.