Recurrent Dictionary Learning for State-Space Models with an Application in Stock Forecasting

Shalini Sharma (1) , Víctor Elvira (2) , Emilie Chouzenoux (3) , Angshul Majumdar (1)

1 IIIT-Delhi - Indraprastha Institute of Information Technology [New Delhi]
2 School of Mathematics - University of Edinburgh
3 OPIS - OPtimisation Imagerie et Santé

In this work, we introduce a new modeling and inferential tool for dynamical processing of time series. The approach is called recurrent dictionary learning (RDL). The proposed model reads as a linear Gaussian Markovian state-space model involving two linear operators, the state evolution and the observation matrices, that we assumed to be unknown. These two unknown operators (that can be seen interpreted as dictionaries) and the sequence of hidden states are jointly learnt via an expectation-maximization algorithm. The RDL model gathers several advantages, namely online processing, probabilistic inference, and a high model expressiveness which is usually typical of neural networks. RDL is particularly well suited for stock forecasting. Its performance is illustrated on two problems: next day forecasting (regression problem) and next day trading (classification problem), given past stock market observations. Experimental results show that our proposed method excels over state-of-the-art stock analysis models such as CNN-TA, MFNN, and LSTM.

Mots clés: Stock Forecasting; Recurrent dictionary learning; Kalman filter; expectation-minimization; dynamical modeling; uncertainty quantification

Domaines: Traitement du signal et de l'image [eess.SP]; Économie et finance quantitative [q-fin]; Intelligence artificielle [cs.AI]

Shalini Sharma, Víctor Elvira, Emilie Chouzenoux, Angshul Majumdar. Recurrent Dictionary Learning for State-Space Models with an Application in Stock Forecasting. Neurocomputing, In press. ⟨hal-03184841⟩