Multi-Factor Stock Selection Strategy Based on Network Sentiment Analysis

Daiyu Qian, Junru Wang

2024

Abstract

With the development of the Internet, how people communicate and express their emotions online is more diverse and timely. It has also become easier to access relevant news information in investor networks, and sentiment can play an important role in the stock market, and investors’ mood swings can affect the direction of the stock market. The general market sentiment is expressed by panic, greed, and uncertainty among investors. For many individual investors, it is easy to follow the market sentiment and make irrational investments without a professional knowledge background. Based on the previous research results, this paper constructs an investor sentiment indicator by using the method of computer text sentiment analysis and further explores its impact on stock returns. This paper uses the method based on text sentiment analysis to construct the investor sentiment index, and the detailed steps are to first scrape the relevant stock comments from the network directly through the Python crawler, and then use the jieba package to clean the crawled data and extract the weight of keywords. Then, the sentiment index is constructed, which is converted into a sentiment index by collecting the keywords in the comments. Finally, this paper uses the sentiment index as the main factor to calculate the OSL regression to analyze the excess return of stocks.

Paper Citation

in Harvard Style

Qian D. and Wang J. (2024). Multi-Factor Stock Selection Strategy Based on Network Sentiment Analysis. In Proceedings of the 1st International Conference on Data Science and Engineering - Volume 1: ICDSE; ISBN 978-989-758-690-3, SciTePress, pages 142-147. DOI: 10.5220/0012829900004547

in Bibtex Style

@conference{icdse24,
author={Daiyu Qian and Junru Wang},
title={Multi-Factor Stock Selection Strategy Based on Network Sentiment Analysis},
booktitle={Proceedings of the 1st International Conference on Data Science and Engineering - Volume 1: ICDSE},
year={2024},
pages={142-147},
publisher={SciTePress},
organization={INSTICC},
doi={10.5220/0012829900004547},
isbn={978-989-758-690-3},
}

in EndNote Style

TY - CONF

JO - Proceedings of the 1st International Conference on Data Science and Engineering - Volume 1: ICDSE
TI - Multi-Factor Stock Selection Strategy Based on Network Sentiment Analysis
SN - 978-989-758-690-3
AU - Qian D.
AU - Wang J.
PY - 2024
SP - 142
EP - 147
DO - 10.5220/0012829900004547
PB - SciTePress