Figure - uploaded by Huthaifa Alqaralleh
Content may be subject to copyright.
Source publication
This study is set out to model and forecast the cryptocurrency market by concentrating on several stylized features of cryptocurrencies. The results of this study assert the presence of an inherently nonlinear mean-reverting process, leading to the presence of asymmetry in the considered return series. Consequently, nonlinear GARCH-type models taki...
Contexts in source publication
Context 1
... results obtained from the nonlinear GARCH models (Note 1) are summarised in Panel A of Table 3. We computed the robust standard errors to obtain robust inferences about the estimated models. ...Similar publications
Background: Based on the static mean-variance portfolio optimisation theory, investors will choose the portfolio with the highest Sharpe ratio to achieve a higher expected utility. However, the traditional Sharpe ratio only accounts for the first two moments of return distributions, which can lead to false portfolio performance diagnostics with the...
Citations
... Besides, the volatility of Bitcoin diminishes when the investors are optimistic about the stock market related to the greater U.S. economic policy uncertainty. Alqaralleh et al. (2020) analyze the potential stylized characteristics of digital currencies. They reveal the existence of nonlinear mean-reverting process. ...
In this paper, we attempt to understand and identify the cyclical fluctuations in cryptocurrency markets. To this end, we apply the Markov-Switching approach on daily prices of 17 selected digital currencies. This model allows us to capture the nonlinear structure in cryptocurrencies’ prices. The empirical results clearly show potential difference(s) among digital currencies when they react to the varying levels of the pandemic's severity. The existence of two distinguishable states and each state seems to be characterized by different features of market cycle’s phase for each cryptocurrency. So, the Covid19 pandemic affects asymmetrically the different market phases of digital currencies. Such findings can have insightful portfolios implications.
... Cryptocurrency prices are output of this complex system. Cryptocurrency prices exhibit high level of nonlinearity, uncertainty and volatility (Chaim and Laurini 2019;Alqaralleh et al. 2020). Therefore, prediction of cryptocurrency prices is very difficult (Mezquita et al. 2022). ...
This issue is dedicated to the memory of Prof. Tenreiro Machado.
https://dergipark.org.tr/en/pub/chaos/issue/64884
... Cryptocurrency prices are output of this complex system. Cryptocurrency prices exhibit high level of nonlinearity, uncertainty and volatility (Chaim and Laurini 2019;Alqaralleh et al. 2020). Therefore, prediction of cryptocurrency prices is very difficult (Mezquita et al. 2022). ...
Cryptocurrencies are new kinds of electronic currencies based on communication technologies. These currencies have attracted the attention of investors. However, cryptocurrencies are very volatile and unpredictable. For investors, it is very difficult to make investment decisions in cryptocurrency market. Therefore, revealing changes in the dynamics of cryptocurrencies are valuable for investors. Bitcoin is the most popular and representative cryptocurrency in cryptocurrency market. In this study how dynamical properties of Bitcoin changed through time is analyzed with recurrence quantification analysis (RQA). RQA is a pattern recognition-based time series analysis method that reveals dynamics of the time series by calculating some metrics called RQA measures. This method has been successfully applied to nonlinear, nonstationary, short and chaotic time series and do not assume a statistical model. RQA can reveal important properties of time series data such as determinism, laminarity, stability, randomness, regularity and complexity. By using sliding window RQA we show that in 2021 RQA measures for Bitcoin prices collapse and Bitcoin becomes more unpredictable, more random, more unstable, more irregular and less complex. Therefore, dynamics and stability of the Bitcoin prices significantly changed in 2021.
... So far, it has not found application with distributional neural networks. However, it is often used in many other areas, e.g., energy and finance markets or medicine [56][57][58][59] and also in electricity price forecasting using regression frameworks [60][61][62] . Based on Figure 3 we suspect that the Johnson's SU is more suitable for modeling the electricity prices than the normal. ...
We present a novel approach to probabilistic electricity price forecasting (EPF) which utilizes distributional artificial neural networks. The novel network structure for EPF is based on a regularized distributional multilayer perceptron (DMLP) which contains a probability layer. Using the TensorFlow Probability framework, the neural network's output is defined to be a distribution, either normal or potentially skewed and heavy-tailed Johnson's SU (JSU). The method is compared against state-of-the-art benchmarks in a forecasting study. The study comprises forecasting involving day-ahead electricity prices in the German market. The results show evidence of the importance of higher moments when modeling electricity prices.
... Their results suggested that a misinterpretation among market participants can cause cryptocurrency markets to be relatively illiquid, thus leading to extreme price movements. Alqaralleh et al. (2020) pursued a different approach in assessing market efficiency and tail behaviour. Their nonlinear GARCH-type models consider the distributions of innovations that capture skewness, kurtosis and heavy tails, providing excellent tools for modelling cryptocurrency returns. ...
Purpose
Research on price extremes and overreactions as potential violations of market efficiency has a long tradition in investment literature. Arguably, very few studies to date have addressed this issue in cryptocurrencies trading. The purpose of this paper is to consider the extreme value modelling for forecasting COVID-19 effects on cryptocoin markets. Additionally, this paper examines the importance of technical trading indicators in predicting the extreme price behaviour of cryptocurrencies.
Design/methodology/approach
This paper decomposes the daily-time series returns of four cryptocurrency returns into potential maximum gains (PMGs) and potential maximum losses (PMLs) at first and then tests their lead–lag relations under an econometric framework. This paper also investigates the non-random properties of cryptocoins by computing the incremental explanatory power of PML–PMG modelling with technical trading indicators controlled. Besides, this paper executes an event study to identify significant changes caused by COVID-19-related events, which is capable of analysing the cryptocoin market overreactions.
Findings
The findings of this paper produce the evidence of both market overreactions and trend persistence in the potential gains and losses from coins trading. Extreme price behaviour explains volatility and price trends in crypto markets before and after the outbreak of a pandemic that substantiate the non-random walk behaviour of crypto returns. The presence of technical trading indicators as control variables in the extreme value regressions significantly improves the predictive power of models. COVID-19 crisis affects the market efficiency of cryptocurrencies that improves the usefulness of extreme value predictions with technical analysis.
Research limitations/implications
This paper strongly supports for the robustness of technical trading strategies in cryptocurrency markets. However, the “beast is moving quick” and uncertainty as to the new normalcy about the post-COVID-19 world puts constraint on making best predictions.
Practical implications
The paper contributes substantially to our understanding of the pricing efficiency of cryptocurrency markets after the COVID-19 outbreak. The findings of continuing return predictability and price volatility during COVID-19 show that profitable investment opportunities for cryptocoin traders are prevailing in pandemic times.
Originality/value
The paper is unique to understand extreme return reversals behaviour of cryptocurrency markets regarding events related to COVID-19 breakout.
... However, evaluating the performances of GARCH models in an out-of-sample context is a more valuable and trusted approach, in practice. Hence, recently, the studies by Trucíos [24], Troster et al. [25], Alqaralleh et al. [26], Köchling et al. [27], and Cerqueti et al. [28] have used a total number of 99, 40,16,22, and 36 GARCH-type models for out-of-sample comparisons, respectively. ...
... In the light of the results, it can be concluded that heavy-tailed GARCH models always lead to better performance in both out-of-sample and goodness-offit contexts and, among them, CGARCH with ged distribution outperforms the other models for out-of-sample performance. Alqaralleh et al. [26] took five cryptocurrencies, including Bitcoin, for an analysis of volatility to make a comparison between nonlinear GARCH-type models. They found that nonlinear GARCH-type models that considered distinctive properties of the returns, such as skewness, kurtosis, and heavier tails using different distributions, offered a better forecasting performance than artificial neural networks. ...
Modelling the volatility of Bitcoin, the cryptocurrency with the largest market share, has recently attracted considerable attention from researchers, practitioners and investors in financial markets and portfolio management. For this purpose, a wide variety of GARCH-type models have been employed. However, there is no consensus in the literature on which specification arising from the volatility equation and the assumed error distribution is better in an out-of-sample performance. This study tries to fill this gap by comparing the forecasting performances of 110 GARCH-type models for Bitcoin volatility. Furthermore, it proposes a new combining method based on support vector machines (SVM). This method effectively selects the set of superior models to perform meta-learning. The results indicate that the best performing GARCH specification depends on the loss function chosen, and the proposed method leads to more accurate volatility forecasts than those of the best GARCH-type models and other combining methods investigated.
The risk-return relationship is of fundamental significance in the field of economics and finance. It is used to structure investment strategies, allocate resources, as well as assist in the construction of policy and regulatory frameworks. The accurate forecast of the risk-return relationship ensures sound financial decisions, whereas an inaccurate one can underestimate risk and thus lead to losses. The GARCH-M approach is one of the foremost models used in South African literature to investigate the risk-return relationship. This study made a novel and significant contribution, on a local and international level, as it was the first study to investigate GARCH-M type models with different innovation distributions. This study analyzed the JSE ALSI returns of the South African market for the sample period from 05 October 2004 to 05 October 2021. Results revealed that the EGARCH (1, 1)-M with the Skewed Student-t distribution (Skew-t) is optimal relative to the standard GARCH, APARCH and GJR. However, the EGARCH-M Skew-t failed to capture the financial data's asymmetric, volatile and random nature. To improve forecast accuracy, this study applied different nonnormal innovation distributions: the Pearson Type Ⅳ distribution (PIVD), Generalized Extreme Value distribution (GEVD), Generalized Pareto distribution (GPD) and Stable. Model diagnostics revealed that the nonnormal innovation distributions adequately captured asymmetry. The Value at Risk and backtesting procedure found that the PIVD, followed by Stable, outperformed the Extreme Value Theory distributions (GEVD and GPD). Thus, investors, risk managers and policymakers would opt to use the EGARCH-M in combination with the PIVD when modelling the risk-return relationship. The main contribution of this study was to confirm that applying GARCH type models with the conventional and normal type distributions, to a volatile emerging market, is considered ineffective. Therefore, this study recommended the exploration of other innovation distributions, that were not included in the scope of this study, for future research purposes.
The inconsistency of growth rate in industrial and its electricity consumption in China has attracted the global attention. This study uses a novel quantile-on-quantile approach to reveal the complicated and seemly inconsistent relationship between China's industrial growth and its electricity consumption from 1995 to 2017. Empirical results show: (1) generally, a positive correlation exists between industrial growth and its electricity consumption. However, the strength of that correlation depends highly on the level of industrial electricity consumption and industrial development, thereby causing variations in the relationships at different periods and regions. (2) During the middle period of industrialization, a strong relationship is observed. Then, it gradually weakened with the steady growth of industry and high electricity consumption. (3) From a regional perspective, the positive correlation in eastern China changed from strong to weak, while a “high-low-high” trend transpired in northeastern China. The relationship was stable in central China during the sample period. (4) The main reason for the above results is the different characteristics of the industrial transfer and industrial structure upgrading across regions.
