What is it about?
Stock market forecasting is a multi-billion dollar industry; yet, curiously, little has been published about the effectiveness of automated stock forecasting strategies. It has been hypothesized that markets are “information efficient,” which ought to make any forecasting attempts equivalent to random guessing. Here, we have tried to refute this hypothesis and shed light on the opaque world of automated forecasting using various algorithms and data sources.
Featured Image
Why is it important?
In this paper, we do a survey of and implement a broad range of strategies and algorithms for the short term (i.e., next day) forecasting. The results of this research demonstrate the strong predictive power of sentiment analysis combined with specific classification algorithms. Lastly, we also provide an easy-to-understand framework for structuring the forecasting activity of independent interest to economists and data scientists.
Perspectives
We hope that this article helps economists grasp the technical terms thrown around by data scientists while also understanding the limitations and capabilities of automated forecasting. At the same time, we think that this paper could serve as an approachable introduction to stock forecasting for data scientists.
Dr. Sanjay Singh
Manipal Institute of Technology, Manipal
On a personal note, working on this paper with my co-authors was a memorable experience. It served as a launchpad for my intellectual curiosity which has led me to pursue a career in research. I consider myself fortunate for this period of exploration that I was privileged to have and wish that all students get to experience something similar.
Mansoor Ahmed-Rengers
University of Cambridge
Read the Original
This page is a summary of: Short Term Firm-Specific Stock Forecasting with BDI Framework, Computational Economics, July 2019, Springer Science + Business Media,
DOI: 10.1007/s10614-019-09911-0.
You can read the full text:
Contributors
The following have contributed to this page
