Impact of COVID-19 Pandemic News on the Cryptocurrency Market and Gold Returns: A Quantile-on-Quantile Regression Analysis
Round 1
Reviewer 1 Report
The paper investigated the relationship between the RavenPack news-based index associated with coronavirus outbreak and returns of two commodities, Bitcoin and gold. Authors found that both commodities, Bitcoin and gold, can serve as a hedge against pandemic-related news.
The quantile-on-quantile method is an adequate statistical method for this data. The result and method are great with this data.
Author Response
Comment: The quantile-on-quantile method is an adequate statistical method for this data. The result and method are great with this data.
Reply: Thank you so much, we are glad to receive such a positive comment.
Author Response File: 
Author Response.pdf
Reviewer 2 Report
The topic of the manuscript is interesting and meets the scope of a journal.
Although there are some flaws, which must be addressed.
- I do not agree, that taking only Bitcoin into the research allows to draw conclusion about the whole cryptocurrency market. Although some of them show correlation in performance (up and downs) it is not unconditionally.
- The data used for a research, namely HI and FNI are a quite controversial and may impose some bias to the results. At least some additional robustness checks are required.
- The developed model fairly works on the past data. Additional argumentation why it should be useful in other situations/other datasets should be provided, as now soem may think about the situation there model is fit for particular data.
Author Response
Comment 1: The topic of the manuscript is interesting and meets the scope of a journal.
Reply: Thank you so much, we appreciate it.
============================
Comment 2: I do not agree, that taking only Bitcoin into the research allows to draw conclusion about the whole crypto-currency market. Although some of them show correlation in performance (up and downs) it is not unconditionally.
Reply: Thank you for your comments. We agree that taking only Bitcoin into the research will not allow to draw conclusion about the whole cryptocurrency market, but we focus our attention on studying Bitcoin as it is well known that it is the largest-capped most prominent digital currency. We add a statement in the conclusion of this paper, which indicate that the idea in this research can be extended to investigate the relationship between the RavenPack news-based index associated with coronavirus outbreak and other cryptocurrencies.
============================
Comment 3: The data used for a research, namely HI and FNI are a quite controversial and may impose some bias to the results. At least some additional robustness checks are required.
Reply: Thank you for your comments. We absolutely agree that the HI and FNI indices have different effects on both Gold and Bitcoin markets in comparison with other indices. In fact, one of the main objectives of this research is to compare theses effects on both commodities (Bitcoin and Gold). The statistical results suggest that the different levels of media-induced hype and fake news affect Bitcoin’s and Gold returns asymmetrically and both (Bitcoin and Gold) act as a hedge against the extreme levels of media-induced hype and fake news. We believe that these findings are interesting which can be investigated in further research projects.
============================
Comment 4: The developed model fairly works on the past data. Additional argumentation why it should be useful in other situations/other datasets should be provided, as now soem may think about the situation there model is fit for particular data.
Reply: Thank you for your comments. Traditionally, researchers use the available past data and try to develop a model that can fairly works with this data. In this research, we do the same and show that the use of the OLS and the quantile regression to estimate the parameters in financial data cases can distort the results significantly, whereas the estimates based on the quantile-on-quantile regression model is the best. We believe that our findings may be helpful for different stakeholders in understanding the cryptocurrency dynamics and in making better investment decisions, especially under adverse conditions and during times of uncertain environments such as in the COVID-19 pandemic.
Author Response File: Author Response.pdf
Round 2
Reviewer 2 Report
Dear authors,
Please include more specific explanation about HI and FNI variable suitability for the research. It is not a robust data.
Author Response
We are very grateful to editors and reviewers for their suggestions. We have modified the draft closely following the comments.
In the amended version, we removed all results regarding these indices. We are planning to study the HI and FNI variables in a near-future research project making sure that the data is robust as suggested by the reviewer.
We hope that these changes will facilitate the decision to publish this article.
Author Response File: Author Response.pdf
Round 3
Reviewer 2 Report
Now I do not have serious objections for publication of this manuscript
