Deep learning approaches for stock market forecasting.

What is it about?

Stock market forecasting is a challenging task as it exhibits highly non linear, volatile and chaotic patterns. With recent success of Deep Learning (DL) in different domains, it has also paved its way in the finance domain. DL methods have an edge over statistical approaches. This work categorizes DL models and reviews the work employed since the last five years for stock market forecasting.

Featured Image

Photo by Nick Chong on Unsplash

Photo by Nick Chong on Unsplash

Why is it important?

1. Synchronizing researchers: The work aims to bring together researchers who are working in the FinTech domain and provide them with an overview of recent progress. This suggests that the work intends to bridge any gaps in knowledge and ensure that researchers are up to date with the latest developments in the field. 2. Comprehensive analysis: The work offers a thorough examination of various deep learning models, features, datasets, and evaluation parameters. By covering these aspects comprehensively, it aims to provide a holistic understanding of the different components involved in FinTech research. This analysis is not only targeted at experienced researchers but also includes information that beginners can benefit from. 3. Identification of advantages and disadvantages: The work highlights both the advantages and disadvantages of the deep learning models, features, datasets, and evaluation parameters discussed. By doing so, it aims to shed light on the strengths and weaknesses of different approaches and methodologies. This analysis helps to identify potential research gaps that can be explored in future studies. 4. Mention of data collection sources: The work also mentions various sources of data collection. This indicates that it covers the different ways in which researchers can obtain the necessary data for their FinTech projects. By providing this information, the work aims to assist researchers in understanding and accessing relevant data sources for their research. In summary, this work serves as a comprehensive resource that brings together researchers in FinTech, provides a detailed analysis of deep learning models, features, datasets, and evaluation parameters, highlights advantages and disadvantages, and offers information on data collection sources. Its objective is to enhance knowledge sharing, support both beginners and experienced researchers, and identify potential research gaps in the FinTech domain.

Perspectives