Data-driven Neural Architecture Learning For Financial Time-series Forecasting
This work addresses the problem of accurate forecasting for financial analysts by improving prediction in nonstationary and nonlinear time-series data, though it appears incremental as it builds on existing neural architecture learning methods.
The paper tackles financial time-series forecasting by proposing a data-driven neural architecture learning algorithm that adaptively learns a mapping function using Generalized Operational Perceptrons, addressing imbalanced data with a modified objective function. Experiments on a large-scale Limit Order Book dataset show it outperforms related algorithms, including tensor-based methods.
Forecasting based on financial time-series is a challenging task since most real-world data exhibits nonstationary property and nonlinear dependencies. In addition, different data modalities often embed different nonlinear relationships which are difficult to capture by human-designed models. To tackle the supervised learning task in financial time-series prediction, we propose the application of a recently formulated algorithm that adaptively learns a mapping function, realized by a heterogeneous neural architecture composing of Generalized Operational Perceptron, given a set of labeled data. With a modified objective function, the proposed algorithm can accommodate the frequently observed imbalanced data distribution problem. Experiments on a large-scale Limit Order Book dataset demonstrate that the proposed algorithm outperforms related algorithms, including tensor-based methods which have access to a broader set of input information.