The Efficiency of Value-at-Risk Models during Extreme Market Stress in Cryptocurrencies

Round 1

Reviewer 1 Report

1.                In the introduction, it should be explicitly shown all relevant data related to a stock market crash, leading to an influx of funds, further stress to the crypto currency market, rapid changes in monetary policies in the US impacting the cryptocurrency market, a collapse of speculative bubbles in the cryptocurrency market.  Thus, gap research can be conveyed so that the main research questions can be supported.

2.                Is this research topic original? Dispositioned with previous research that is relevant to this topic, and where the development or novelty of this research is a contribution in the field of financial technology. Because it seems that this study focuses more on evaluating efficiency through historical simulations with the VaR method, Delta Normal VaR and the Monte Carlo Simulation VaR applied to the top five largest cryptocurrencies by market capitalization (BTC, ETH, BNB, ADA, and XRP).

3.                In the research methodology, it can be made more specifically, especially about the level of analysis used in this study. Does this fit the purpose of the study?

4.                In the results of the study, please sharpen what important findings are obtained so that the results of the test can be statistically interpreted firmly and placed in the discussion chapter.

5.                There must be further consequences of the conclusions written in the form of managerial implications and theoretical implications.

6.                It can be added to update the latest research research in the last three years.

Author Response

Dear Reviewers,

We extend our sincere appreciation for your efforts in reviewing our paper. Your insightful feedback and recommendations have been of immense assistance in enhancing the overall quality of our work. In response to your suggestions, we have made the necessary modifications to the paper. Additionally, we would like to take the opportunity to address and clarify any questions or points raised in your comments. Please see the attachment.

1. In the introduction, it should be explicitly shown all relevant data related to a stock market crash, Suggestions for Authors leading to an influx of funds, further stress to the crypto currency market, rapid changes in monetary policies in the US impacting the cryptocurrency market, a collapse of speculative bubbles in the cryptocurrency market. Thus, gap research can be conveyed so that the main research questions can be supported.

We have included explicit, relevant data to the introduction in accordance with your suggestion, in line 50 - 67.

“There is evidence from BTC’s market capitalization which has reached its peak of $1.28 trillion on November 9, 2021. It was approximately 10 times, at $117.15 billion, higher than its value on March 31, 2020. Despite of receiving a large amount of fund flow, the study results from Lahmiri & Bekiros [‎17] and Vojtko & Cisár [‎4] suggest that the cryptocurrency market exhibits a higher level of risk due to the pandemic. Additionally, the escalating conflict between Russia and Ukraine has added further stress to the cryptocurrency market resulting in more volatility [‎18]. Khalfaoui et al. [‎19] explained that there were sell-offs from large holders in response to the war which resulted in a decline in prices. Furthermore, there have been rapid changes in monetary policies in the developed economies, for instance the FED interest rates raise from 0%-0.25% in March 2020 to 4.25%-4.5% in December 2022 [‎20], which have also had an impact on the cryptocurrency market [‎‎21,‎‎22]. Moreover, the study results of Yu & Chen [‎23] indicate that China regulatory ban in cryptocurrency mining and transactions in May 2021. Additionally, it has been suggested that there has been a collapse of speculative bubbles in the cryptocurrency market during the period of our study [‎24-‎27]. Bazán-Palomino [‎24] discovered that the collapse of the bubble resulted in a higher market volatility and expected shortfall. During this period, the market stress has led to a maximum drawdown of BTC was at -76.71%.”

2. Is this research topic original? Dispositioned with previous research that is relevant to this topic, and where the development or novelty of this research is a contribution in the field of financial technology. Because it seems that this study focuses more on evaluating efficiency through historical simulations with the VaR method, Delta Normal VaR and the Monte Carlo Simulation VaR applied to the top five largest cryptocurrencies by market capitalization (BTC, ETH, BNB, ADA, and XRP).

This research is an important contribution to the field because it empirically provides practical information on risk modelling with VaR models for cryptocurrencies market risk during a market stress which can potentially cause VaR models to lose its accuracy.

Other recent research in this field focuses on Expected Shortfall which is an extreme risk measure for market turbulence. According to our results the Delta Normal VaR and Monte Carlo Simulation VaR tends to overestimate the risk. Therefore, the Expected Shortfall values will surely overestimate risk as well since it is an extreme risk measure which considers the tail losses.

3. In the research methodology, it can be made more specifically, especially about the level of analysis used in this study. Does this fit the purpose of the study?

The purpose of the study has been stated in line 72-79

“The market stress of cryptocurrencies may result in the loss of accuracy of VaR models, which have been criticized for their usage in times of financial turmoil. Therefore, in this paper, we will evaluate the efficiency of the Historical Simulation VaR (HS VaR), Delta Normal VaR (DN VaR) and the Monte Carlo Simulation VaR (MC VaR) when they are applied on the top five cryptocurrencies BTC, ETH, BNB, ADA, and XRP. The time horizon of analysis is between March 31, 2020, and December 25, 2022, a period during which the cryptocurrency market experienced a large inflow of funds from investors and market stress.”

Explanations of the research methodology was improved regarding your suggestion in the backtesting methodology subsection, line 287-303.

“An efficient VaR model must be capable of accurately predicting potential losses and maintain its robustness over varying periods of time. To assess the efficiency of a model, backtesting must be employed. Zhang and Nadarajah [‎98] have categorized backtesting methods into 4 categories, including unconditional tests, conditional tests, independence property tests, and other methods.

The unconditional method focuses on the violations to determine accuracy or adequacy of a VaR model, but it cannot detect patterns in its performance. Hence, testing the independence properties of a VaR model's performance is crucial as a robust model should perform equally throughout the time horizon. Therefore, in this study, both the independence property approach and conditional approach are utilized to further evaluate the efficiency of VaR models.

This paper employs the unconditional test method including Kupiec's POF test and Kupiec's TUFF test. For independence property test, the Independence Test [‎99] is used to test the robustness of the VaR models. Finally, the Kupiec’s POF tests and Independence test are combined for the Christoffersen's Interval Forecast test, which is a conditional coverage test that examines the accuracy and robustness of the VaR models simultaneously.”

4. In the results of the study, please sharpen what important findings are obtained so that the results of the test can be statistically interpreted firmly and placed in the discussion chapter.

We have sharpened the important findings in the results section following your suggestion:

Line 382-384

“Overall, the results from these tests suggest that the HS VaR model is the appropriate during market stress, as it demonstrates efficiency in terms of accuracy and robustness.”

Line 411-414

“In conclusion, the DN VaR model loss some accuracy due to the market stress but the model still maintains its robustness. The overestimation of the risk provides adequate VaR measure which can benefit conservative risk management strategies and capital reserves planning.”

Line 444-448

“In summary, the tests indicate that MC VaR model experiences a reduction in accuracy during market stress. However, the model still shows stability of its performance. Fur-thermore, it mainly provides overestimation, making it suitable for conservative risk management practices.”

5. There must be further consequences of the conclusions written in the form of managerial implications and theoretical implications.

An explanation of our findings for practical use in risk management and a theoretical perspective in line 483-492.

“For Managerial implications, it is recommended that the HS VaR model should be employed for active risk management strategies during market stress, as the results from unconditional tests indicate that the model consistently provides balanced predictions. On the other hand, the DN VaR and MC VaR models are recommended for conservative risk management strategies, as they mostly provide risk overestimations, but still maintain a level of robustness comparable to the HS VaR.

In a theoretical perspective, our results suggest that VaR models should be designed for flexibility to accommodate various distribution shapes as our empirical findings indicate that the DN VaR and MC VaR models, which are based on normal distributions, produce overestimated values.”

6. It can be added to update the latest research in the last three years.

Recent references in the last three years were added:

• Lahmiri, S., & Bekiros, S. (2020). The impact of COVID-19 pandemic upon stability and sequential irregularity of equity and cryptocurrency markets. Chaos, Solitons & Fractals, 138, 109936. doi: 10.1016/j.chaos.2020.109936
• Le, T. H. (2023). Quantile time-frequency connectedness between cryptocurrency volatility and renewable energy volatility during the COVID-19 pandemic and Ukraine-Russia conflicts. Renewable Energy, 202, 613-625. doi: 10.1016/j.renene.2022.11.062
• Almeida, J., & Gonçalves, T. C. (2022). A systematic literature review of volatility and risk management on cryptocurrency investment: A methodological point of view. Risks, 10(5), 107.
• Doumenis, Y., Izadi, J., Dhamdhere, P., Katsikas, E., & Koufopoulos, D. (2021). A critical analysis of volatility surprise in Bitcoin cryptocurrency and other financial assets. Risks, 9(11), 207.
• Jiang, Z., Mensi, W., & Yoon, S.-M. (2023). Risks in Major Cryptocurrency Markets: Modeling the Dual Long Memory Property and Structural Breaks. Sustainability, 15(3), 2193.
• Mužić, I., & Gržeta, I. (2022). Expectations of macroeconomic news announcements: Bitcoin vs. Traditional assets. Risks, 10(6), 123.

 

We look forward to your continued feedback and support.

 

Sincerely,

Dr.Danai Likitratcharoen

Author Response File: Author Response.pdf

Reviewer 2 Report

Comments for author File: Comments.pdf

Author Response

Dear Reviewers,

We extend our sincere appreciation for your efforts in reviewing our paper. Your insightful feedback and recommendations have been of immense assistance in enhancing the overall quality of our work. In response to your suggestions, we have made the necessary modifications to the paper. Additionally, we would like to take the opportunity to address and clarify any questions or points raised in your comments. Please see the attachment.

1. Why do the authors limit themselves to VaR only? As a reminder, VaR has been the subject of several criticisms after the subprime crisis. The Expected Shortfall (ES) or conditional VaR or tail loss has been increasingly used following the financial crisis. The properties of ES are better than those of VaR as they encourage, in particular, diversification (Hull, J. (2012). Risk management and financial institutions, Web Site (Vol. 733). John Wiley & Sons. The following paper offers some insights into the issue: Acereda, B., Leon, A., & Mora, J. (2020). Estimating the expected shortfall of cryptocurrencies: An evaluation based on backtesting. Finance Research Letters, 33, 101181.

It is true that VaR is often criticized for its predictability during financial crises. However, our results indicate that the Delta Normal VaR and Monte Carlo Simulation VaR majorly overestimated the risk which means that these two VaR models are adequate. Moreover, the Historical Simulation VaR is found to provide accurate predictions of the losses. Therefore, the Expected Shortfall would further overestimate the risk. Your suggestion has inspired us to include the criticism of VaR in line 104-105.

“However, the study results of Mavani (2020) and Kourouma et al. (2010) show that VaR models are unfit in times of financial crisis.”

2. The authors should also announce the outline of the paper at the end of the introduction. Outline the different sections that should follow.

An outline of the paper has been added at the end of the introduction in line 88-93

“In the next section, we will provide a comprehensive overview of existing literature related to cryptocurrencies’ market risk, the market stress during the time horizon of our research and the underlying blockchain mechanisms of each cryptocurrency. Following, the next section, we provide a detailed explanation of the data, which is used in this paper, the VaR models and backtesting methodology. Lastly, we discuss and summarize the main findings, the managerial and theoretical implications of this paper.”

3. The authors do not motivate the use of the three methods of VaR estimation. Why limit themselves to constant volatility models. It would be interesting to estimate VaR assuming dynamic volatility. Pafka, S., & Kondor, I. (2001). Evaluating the RiskMetrics methodology in measuring volatility and Value-at-Risk in financial markets. Physica A: Statistical Mechanics and its Applications, 299(1-2), 305-310. Mina, J., & Xiao, J. Y. (2001). Return to RiskMetrics: the evolution of a standard. RiskMetrics Group, 1, 1-11.

We wanted to test the three models which have been proven to be robust in the past in which our result have proven that the Historical Simulation VaR model provides an accurate estimation of the downside risk. Moreover, we have found that the Delta Normal VaR and Monte Carlo Simulation VaR provide an adequate VaR measure since it overestimates the risk as shown by number of violations which is significantly lower than the suggested number of violations at 95% and 90% confidence level.

4. The literature has also proposed several types of volatility: deterministic volatility, random but conditionally deterministic volatility, and stochastic volatility (Hull and White, 1987). The authors should address these very important concepts. Hull, J., & White, A. (1987). The pricing of options on assets with stochastic volatilities. The journal of finance, 42(2), 281-300.

An explanation has been added in line 277-285

“Equation (5) presumes a fixed volatility, which does not correspond to the realities of financial markets. For instance, cryptocurrency markets are notorious for their high volatility and can experience rapid fluctuations due to a multitude of factors. To account for this, the constant volatility assumption in equation (5) can be relaxed and replaced with models such as GARCH or stochastic volatility models, such as the Hull and White model [‎95] or the CIR model [‎96] introduced by Heston [‎97], which capture the underlying price's stochastic volatility. Nevertheless, the scope of this paper does not encompass such models and merely compares three models. As a result, this paper will assume a constant volatility in equation (5).”

5. Was the estimation of geometric Brownian motion in the Monte Carlo simulation performed in the real or risk-neutral environment? Please explain in relation to equation 5.

An explanation has been added in line 269-274

“This equation has been found to perform well under both real-world and risk-neutral measures. The choice between the two measures depends on the objective of the VaR calculation, whether to reflect real market conditions or a risk-neutral perspective. In this paper, the model will be presented using the risk-neutral measure. As a result, equation (5) is suitable for use in the Monte Carlo simulation and a price return process can be derived using Itô's Lemma as follows.”

 

We look forward to your continued feedback and support.

 

Sincerely,

Dr.Danai Likitratcharoen

Author Response File: Author Response.pdf

Round 2

Reviewer 1 Report

This article can be recommended to be developed again in terms of its relevance to the characteristics of investors for capital markets in emerging markets

Reviewer 2 Report

The authors have responded satisfactorily to the various questions