Next Article in Journal
Towards Fair AI: Mitigating Bias in Credit Decisions—A Systematic Literature Review
Next Article in Special Issue
Interconnectedness of Stock Indices in African Economies Under Financial, Health, and Political Crises
Previous Article in Journal
Nexus Between Fintech Innovations and Liquidity Risk in GCC Banks: The Moderating Role of Bank Size
Previous Article in Special Issue
Modelling Value-at-Risk and Expected Shortfall for a Small Capital Market: Do Fractionally Integrated Models and Regime Shifts Matter?
 
 
Systematic Review
Peer-Review Record

Factors, Forecasts, and Simulations of Volatility in the Stock Market Using Machine Learning

J. Risk Financial Manag. 2025, 18(5), 227; https://doi.org/10.3390/jrfm18050227
by Juan Mansilla-Lopez 1, David Mauricio 2,* and Alejandro Narváez 3
Reviewer 1: Anonymous
Reviewer 2: Anonymous
J. Risk Financial Manag. 2025, 18(5), 227; https://doi.org/10.3390/jrfm18050227
Submission received: 24 February 2025 / Revised: 19 March 2025 / Accepted: 18 April 2025 / Published: 24 April 2025
(This article belongs to the Special Issue Machine Learning-Based Risk Management in Finance and Insurance)

Round 1

Reviewer 1 Report

Comments and Suggestions for Authors

The paper provided a systematic literature review of 102 studies on stock market volatility, consolidating factors, ML algorithms, and simulation models. This fills a critical gap, as prior reviews focused narrowly on forecasting and lacked recent ML advancements. This paper identified 15 factors influencing volatility (grouped into six categories) and 24 ML algorithms, emphasizing hybrid models. This categorization aids researchers in navigating a fragmented field. It highlighted the shift from classical econometric models such as GARCH to ML-based simulations, providing actionable insights for investors and policymakers.

To further strengthen the paper, the following areas should be addressed.

While the paper reviewed existing studies, limited critical engagement with limitations of reviewed studies are provided, for example it did not critically assess the quality, biases, or representativeness of datasets used in the reviewed works.

In addition, traditional models such as GARCH are mentioned but not rigorously compared with ML approaches in terms of computational efficiency, interpretability, or robustness during crises.

Suggestions to integrate generative AI such as ChatGPT are speculative and lack grounding in the reviewed literature. A focus on immediate gaps such as explainability would strengthen recommendations.

Tables listing factors and algorithms may be moved to an appendix to make the paper concise.

 

Author Response

Comments 1: While the paper reviewed existing studies, limited critical engagement with limitations of reviewed studies are provided, for example it did not critically assess the quality, biases, or representativeness of datasets used in the reviewed works.

Response 1: It was addressed on page 11, second paragraph.

Comments 2: In addition, traditional models such as GARCH are mentioned but not rigorously compared with ML approaches in terms of computational efficiency, interpretability, or robustness during crises.

Response 2: Addressed on page 11, second paragraph.

Comments 3: Suggestions to integrate generative AI such as ChatGPT are speculative and lack grounding in the reviewed literature. A focus on immediate gaps such as explainability would  strengthen recommendations.

Response 3: The mention of Generative AI (ChatGPT) has been removed, and the gap regarding explainability in the stock market has been included on page 12.

Comments 4: Tables listing factors and algorithms may be moved to an appendix to make the paper concise.

Response 4: It was addressed from page 16 to 19.

Author Response File: Author Response.docx

Reviewer 2 Report

Comments and Suggestions for Authors

Dear Authors,

Thank you for your diligent efforts in the preparation of this paper; your hard work has greatly enhanced its quality and impact in this manuscript.

The paper is generally well written and structured, and authors picked an interesting topic related to the technological techniques used in the stock markets. It focuses on the examination of factors, forecasts and simulations of the stock market by using machine learning. This manuscript is considered as a contribution to the field of financial markets and provides valuable insights into the use of ML algorithms for stock market volatility analysis. The paper is well structured, the methodology is clear, the literature review is comprehensive.

The authors used the econometric approach based on the Panel VAR techniques to analyze this work.

Although the efforts made by authors, in my opinion the paper has some shortcomings in regards of the following:

  • The paper focuses on the factors, forecasts and simulations of stock market volatility but it excludes other applications such as the predictions of stock prices.
  • I think that the paper presents many details purely technical, and it seems that it is very dense. I suggest authors to simplify some sections and provide more explanatory context that could improve the readability of the manuscript.
  • Some of the figures and tables are not well-integrated into the manuscript. Authors should ensure that all figures and tables are clearly referenced.
  • Authors presents a strong variety of ML algorithms, but they didn’t provide a definitive conclusion about the bet efficient algorithm for factors, forecasts and simulations of stock market volatility.
  • Authors briefly mention limitations. They could benefit from a more thorough discussion of the limitations and their implications for their findings.
  • In the future research, it is suggested to address the identified weakness and to explore new algorithms for improved volatility forecasting.

Author Response

Comments 1: The paper focuses on the factors, forecasts and simulations of stock market volatility but it excludes other applications such as the predictions of stock prices.

Response 1: The research is limited to articles that focus solely on volatility and its measurement (a measure of stock market risk); applications related to stock price predictions aim to determine their future monetary value.

Comments 2: I think that the paper presents many details purely technical, and it seems that it is very dense. I suggest authors to simplify some sections and provide more explanatory context that could improve the readability of the manuscript.

Response 2: It was addressed from page 16 to 19.

Comments 3: Some of the figures and tables are not well-integrated into the manuscript. Authors shouldensure that all figures and tables are clearly referenced. 

Response 3: It was addressed, the table numbering has been corrected and annexes have been considered

Comments 4: Authors presents a strong variety of ML algorithms, but they didn’t provide a definitive conclusion about the bet efficient algorithm for factors, forecasts and simulations of stock market volatility.

Response 4: It was addressed on page 11 

Comments 5: Authors briefly mention limitations. They could benefit from a more thorough discussion of the limitations and their implications for their findings.

Response 5: It was addressed on page 13, fifth paragraph. The article is limited to two databases (WOS and Scopus), only articles in English, and a 20-year period. It could be extended to a longer period, include other languages, and incorporate conference papers; however, the considered period is extensive.

Comments 6: In the future research, it is suggested to address the identified weakness and to explore new algorithms for improved volatility forecasting

Response 6: It was addressed on page 12.

Author Response File: Author Response.docx

Round 2

Reviewer 1 Report

Comments and Suggestions for Authors

The paper has made significant progress in addressing the comments provided in the initial review.

Back to TopTop