Review Reports

Round 1

Reviewer 1 Report

Dear Authors,

Here are my comments:

1. Authors are suggested to remove this line from the abstract: "Furthermore, we adopt EUA... analyses."

2. Authors are suggested to remove "EUA (European Union Allowances)" from keywords... Then replace the keyword "volatility-correlation relations" with volatility.

3. Please add reference for first paragraph of introduction section.

4. Authors need to justify the reason for using regime-switching approach in the introduction section. Did previous studies highlighted it ... if so then is there any directives provided ?

5. Overall introduction section need to be more comprehensive.

6. Dataset have COVID period... It is not discussed in the study. For robustness checks of the models... data could be bifurcated into COVID and non COVID period. 

7. Please add future directives add the end of the conclusion section.

8. Recheck all minor grammatical errors if any.

Good Luck !!!

Author Response

Our point-to-point response report is attached. 

Author Response File: Author Response.pdf

Reviewer 2 Report

I read the paper with interest. My only concern is the assumption of the bivariate normal distribution of the innovations in the GARCH process, which were missing in the model description (1)-(6).  I doubt the innovations are Gaussian. Most likely they are heavy-tailed and skewed. I suggest the authors take the bivariate sample innovations and fit bivariate generalized hyperbolic distribution, and thus check how far from normality is the distribution of the GARCH innovations.

Ref. Xiang Deng & Jing Yao (2018) On the property of multivariate generalized hyperbolic distribution and the Stein-type inequality, Communications in Statistics - Theory and Methods, 47:21, 5346-5356, DOI: 10.1080/03610926.2017.1390134

Author Response

Our point-to-point response report is attached. Thank you. 

Author Response File: Author Response.pdf

Reviewer 3 Report

My comments are reported below:

- The Abstract is too long and it must be shortened

- There are several typos in the text: use the free version of Grammarly.com to correct the text

- Not clear what you are forecasting in Tables 7: the returns? the volatilities? Which tests do you use to compare the MAE/MSE. Shortly: section 5.2 must be fully re-written

- In section 5.3 you must add some traditional benchmark portfolio models, like the equally-weighted portfolio, and the Markowitz global minimum variance portfolio. Moreover, Sharpe ratios and other traditional portfolio performance measures must be added.

Author Response

Our point-to=point response report is attached. Thank you. 

Author Response File: Author Response.pdf

Round 2

Reviewer 1 Report

Dear Authors,

Thanks for the revised version. You have did good work.
You have made a comment "To address the issue of the COVID-19 pandemic, one may exclude the data after 2020 and rerun the models. However, running the nonlinear regime-switching model is time-consuming. Given the fact that the Journal gives us only ten days to re-submit the revised paper, we could be unable to include these additional time robustness tests in our paper."

In this regard did authors wrote to the journal requesting more time to revise. I believe journal editorial board will not mind to provide you more time of 1 month or more for revision.

In section 5.2, author(s) have added the following "we construct the equal-weighted energy-carbon for each of the three models" ... In this regard the result may vary with value weighted portfolios ... If it is possible for the author(s) to compare ... please do it ... otherwise include it as future directives. 

Good Luck !!!

Author Response

The point-to-point response report is attached. 

Author Response File: Author Response.pdf

Round 3

Reviewer 1 Report

Good Luck !!!