English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

A linear multi-factor model is one of the most important tools in equityportfolio management. The linear multi-factor models are widely used becausethey can be easily interpreted. However, financial markets are not linear andtheir accuracy is limited. Recently, deep learning methods were proposed topredict stock return in terms of the multi-factor model. Although these methodsperform quite well, they have significant disadvantages such as a lack oftransparency and limitations in the interpretability of the prediction. It isthus difficult for institutional investors to use black-box-type machinelearning techniques in actual investment practice because they should showaccountability to their customers. Consequently, the solution we propose isbased on LSTM with LRP. Specifically, we extend the linear multi-factor modelto be non-linear and time-varying with LSTM. Then, we approximate and linearizethe learned LSTM models by LRP. We call this LSTM+LRP model a deep recurrentfactor model. Finally, we perform an empirical analysis of the Japanese stockmarket and show that our recurrent model has better predictive capability thanthe traditional linear model and fully-connected deep learning methods.

Deep Recurrent Factor Model: Interpretable Non-Linear and Time-Varying Multi-Factor Model