A Hybrid Approach for the Assessment of Risk Spillover to ESG Investment in Financial Networks

Round 1

Reviewer 1 Report

The paper is timely and interesting. However, I have the following comments to further improve the paper.

While the paper has several strengths, there are some limitations that need to be addressed. Firstly, the approach relies heavily on the quality and completeness of the data used to construct financial networks and predict systemic risk. The paper does not provide a detailed discussion of how the authors ensure the quality and completeness of the data used.

Secondly, the approach may oversimplify the complexity of systemic risk outbreaks by focusing only on asset liquidation and not taking into account other types of systemic risk, such as credit or market risk. The authors could have provided a more detailed discussion of the limitations of their approach in capturing systemic risk.

Thirdly, the paper could have provided a more in-depth discussion of the machine learning techniques used to predict systemic risk. The authors could have provided more details about the limitations of these techniques, such as overfitting, underfitting, and the possibility of data bias. Additionally, the authors could have discussed alternative machine learning techniques that could have been used to predict systemic risk.

Finally, while the paper highlights the importance of ESG investments, it does not explore the broader implications of systemic risk on the real economy or social welfare. Therefore, the results may not fully reflect the broader implications of risk contagion in financial systems.

Overall, the paper presents an innovative approach to evaluating risk contagion in financial networks by combining financial network modeling with machine learning techniques. However, the authors could have provided more details about the limitations of their approach and the machine learning techniques used. Additionally, the authors could have discussed the broader implications of systemic risk on the real economy and social welfare.

I would request the author(s) to enhance the writing quality of the paper and ensure it goes through a thorough proofreading process. Check the in-text citations and reference list. 

Author Response

Please see the attachment

Author Response File: Author Response.pdf

Reviewer 2 Report

Journal: Sustainability

Manuscript ID: sustainability-2258017

Type: Article

 

Title: A hybrid approach for the assessment of risk spillover to ESG investment in financial networks

I have carefully revised the manuscript. While it is interesting and well written, I have following comments and suggestions.

1.      Introduction is well-written with a compelling story; however, it contains only 6 references. It is more like commentary than introduction of a scientific study. I suggest adding some recent publications on this topic to improve the introduction in the light of recent studies.

2.      In the introduction sections, please also name the three machine learning tools you will use in the study (If I am not wrong, RF, XGBoost, and DNN are employed).

3.      Please write a concluding statement on the literature review section.

4.      Please further elaborate the data section. I am not sure about the data span and type by reading the following statement between Lines 242-243: “The data set used in this study was obtained from daily observations as of September 2022, encompassing financial conglomerates in the U.S. banking, security, and insurance sector.”

5.      How did the authors match the data, which was obtained using different sources? I did not understand the following statements between Lines 256-257: “In addition, we used ESG investment data associated with these institutions. The data comes from the Bloomberg terminal. We will compare the difference in performance of these institutional portfolios with ESG portfolios”. Furthermore, comparing performance is one of the objectives of this study?

6.      Line 259: The authors stated “In addition to these centrality measures, we have also collected financial information about the institutions in our sample, including total assets, net worth, and return on equity”. Comment: Are these additional variables or the key variables indicated in the top-left part of Figure 1? By looking at the data, why not to adopt logistic regression for binary classification among the machine learning process? Please respond to this comment in the response file only (not in the manuscript at this stage). Thank you.

7.      Please define “abnormal volatility”. Calculating volatility is itself an interesting topic with different measures, such as implied and realized volatility. Indicate clearly, what does “abnormal volatility” mean?

8.      A comparison of performance is presented in Figure 8, there are numerous ways to calculate “returns”. The authors should illustrate the formula to calculate returns?

9.      The conclusion section should include limitations and a future research agenda, extending this study. ESG based diversification and the use of machine and deep learning in finance are popular in recent years. For example, Ali et al (2022) use machine learning tools to fetch tweets and estimate their polarity and subjectivity using two different dictionaries. Similarly, He at al. (2022) discuss corporate social responsibilities and idiosyncratic risk in the light of ESG information disclosure. Khalfaoui et al (2022) shed light on the climate risk and clean energy spillovers, and the impact of uncertainty in US stock markets. Zhang et al. (2023) employed extreme quantile spillovers to understand drivers among clean energy, electricity, and energy metals markets.

10.   In some, if the authors shed light on limitations, future extensions, and applications of the work done in academia/practice by discussing the suggested (or any other) elements and papers, then it will surely enhance the quality as well as the application and citations of this manuscript in the future. Good luck.

Overall, I suggest revised submission of the manuscript before its reconsideration. Good luck.  

 

References:

Ali et al. (2022); DOI: https://doi.org/10.1016/j.ribaf.2022.101768

He at al. (2023); DOI: https://doi.org/10.1016/j.frl.2022.102936

Khalfaoui et al (2022); DOI: https://doi.org/10.1016/j.techfore.2022.122083

Zhang et al (2023); DOI: https://doi.org/10.1016/j.irfa.2022.102474

 

 

 

Author Response

Please see the attachment

Author Response File: Author Response.pdf

Round 2

Reviewer 2 Report

The authors have addressed my comments. Good luck