Figure 1 - uploaded by Yahua Xu
Content may be subject to copyright.
Source publication
This paper reports evidence of intraday return predictability, consisting of both intraday momentum and reversal, in the cryptocurrency market. Using high-frequency price data on Bitcoin from March 3, 2013, to May 31, 2020, it shows that the patterns of intraday return predictability change in the presence of large intraday price jumps, FOMC announ...
Contexts in source publication
Context 1
... calculate the averages trading volume and Amihud (2002) ration in each hourly interval, shown by Fig.1 (a) and Fig.1 (b), respectively. Fig.1(a) provides the visual depiction of the intraday trading volume. ...
Context 2
... calculate the averages trading volume and Amihud (2002) ration in each hourly interval, shown by Fig.1 (a) and Fig.1 (b), respectively. Fig.1(a) provides the visual depiction of the intraday trading volume. ...
Context 3
... calculate the averages trading volume and Amihud (2002) ration in each hourly interval, shown by Fig.1 (a) and Fig.1 (b), respectively. Fig.1(a) provides the visual depiction of the intraday trading volume. ...
Context 4
... also present the intraday pattern of the illiquidity measured by the Amihud measure (Amihud, 2002) which captures the price changes per unit change of the trading volume, presented by Fig.1 (b). As expected, the illiquidity level is negatively correlated with the intraday trading volume. ...
Context 5
... Fig.1(a) represents the average of the intraday trading volume in each hourly interval all over the day. ...
Context 6
... Fig.1(a) represents the average of the intraday trading volume in each hourly interval all over the day. Fig.1(b) shows the average of the illiquidity (i.e., Amihud ratio) in each hourly interval. ...
Context 7
... Figure A.1 about here] Besides, bitcoinity also provides the market shares of Bitcoin trading in various fiat currencies, as shown by Figure A.2. We can see that Bitcoin trades in CNY is ranked in the first place with trading volume of 1.40G BTC and a market share of 80.26%, whereas Bitcoin trades in USD is ranked in the second place with a trading volume of 239M BTC and a market share of 13.74%. ...
Similar publications
The Daniel-Hirshleifer-Sun (DHS) is a three-factor model based on the investors psychology. It supplements the market factors of the CAPM model with two behavioral factors that capture commonalities in mispricing resulting from psychological biases. The DHS method focuses on two psychological biases affecting asset prices: overconfidence and limite...
Citations
... Recently, there has been an outburst in research on intraday price predictability across diverse markets (Huddleston et al, 2023;Wen et al, 2022). Results provide strong evidence that machine learning methods can effectively predict market returns within intraday horizons, beyond what can be explained by transaction costs (Huddleston et al, 2023). ...
Deep reinforcement learning (DRL) is a well-suited approach to financial decision-making, where an agent makes decisions based on its trading strategy developed from market observations. Existing DRL intraday trading strategies mainly use price-based features to construct the state space. They neglect the contextual information related to the position of the strategy, which is an important aspect given the sequential nature of intraday trading. In this study, we propose a novel DRL model for intraday trading that introduces positional features encapsulating the contextual information into its sparse state space. The model is evaluated over an extended period of almost a decade and across various assets including commodities and foreign exchange securities, taking transaction costs into account. The results show a notable performance in terms of profitability and risk-adjusted metrics. The feature importance results show that each feature incorporating contextual information contributes to the overall performance of the model. Additionally, through an exploration of the agent's intraday trading activity, we unveil patterns that substantiate the effectiveness of our proposed model.
... Recently, there has been an outburst in research on intraday price predictability across diverse markets (Huddleston et al. 2023;Wen et al. 2022). Results provide strong evidence that machine learning methods can effectively predict market returns within intraday horizons, beyond what can be explained by transaction costs (Huddleston et al. 2023). ...
Deep reinforcement learning (DRL) is a well-suited approach to financial decision-making, where an agent makes decisions based on its trading strategy developed from market observations. Existing DRL intraday trading strategies mainly use price-based features to construct the state space. They neglect the contextual information related to the position of the strategy, which is an important aspect given the sequential nature of intraday trading. In this study, we propose a novel DRL model for intraday trading that introduces positional features encapsulating the contextual information into its sparse state space. The model is evaluated over an extended period of almost a decade and across various assets including commodities and foreign exchange securities, taking transaction costs into account. The results show a notable performance in terms of profitability and risk-adjusted metrics. The feature importance results show that each feature incorporating contextual information contributes to the overall performance of the model. Additionally, through an exploration of the agent’s intraday trading activity, we unveil patterns that substantiate the effectiveness of our proposed model.
... The increased market acceptance, regulation, and trading modeling also indicate a growing interest in the market. (Bouri et al., 2019(Bouri et al., , 2019Qureshi et al., 2020;Shahzad et al., 2022;Wen et al., 2022;Yousaf et al., 2021), Ahmet Sensoy ( ...
Cryptocurrency's unique features-decentralisation, anonymity, and diversification-have propelled it into the spotlight, attracting both investors and researchers despite its relative youth compared to traditional markets. This study utilizes a systematic mapping approach to examine the current state of cryptocurrency trading research. We are particularly interested in influential variables and technologies involved in cryptocurrency trading systems. By summarizing key findings on data, technology compatibility, and future research directions, this study serves as a starting point for new research activities in the field of cryptocurrency trading.
... Volume and frequency are inversely proportional . Thus, future research topics include detecting price manipulation in cryptocurrencies at the intraday level, using feature selection techniques to improve accuracy, analyzing the impact of hard and soft forks on intraday reversal and momentum, assessing time-based momentum strategies during price crashes and sharp recovery using highfrequency data (Wen et al. 2022), and examining causal patterns between Bitcoin and other cryptocurrencies. Additionally, the volume-return relationship of cryptocurrencies (Balcilar et al. 2017;Yarovaya and Zięba 2022) can be examined with the inclusion of seasonality and volatility effects. ...
... Despite the various perspectives, the primary conclusion of this study is that Bitcoin is inefficient (Kristoufek 2018;Mensi et al. 2019;Nadarajah and Chu 2017;Urquhart 2016) but may achieve weak-form efficiency over time (Chu et al. 2019;Gradojevic et al. 2022). Based on evidence of increased economic value and trading anomalies during the COVID-19 pandemic, Wen et al. (2022) suggest that traders can form a profitable strategy based on historical intra-day returns, leading to an inefficient cryptocurrency market. ...
As the crypto-asset ecosystem matures, the use of high-frequency data has become increasingly common in decentralized finance literature. Using bibliometric analysis, we characterize the existing cryptocurrency literature that employs high-frequency data. We highlighted the most influential authors, articles, and journals based on 189 articles from the Scopus database from 2015 to 2022. This approach enables us
to identify emerging trends and research hotspots with the aid of co-citation and car- tographic analyses. It shows knowledge expansion through authors’ collaboration
in cryptocurrency research with co-authorship analysis. We identify four major streams of research: (i) return prediction and measurement of cryptocurrency volatility, (ii) (in) efficiency of cryptocurrencies, (iii) price dynamics and bubbles in cryptocurrencies, and (iv) the diversification, safe haven, and hedging properties of Bitcoin. We conclude that highly traded cryptocurrencies’ investment features and economic outcomes are analyzed predominantly on a tick-by-tick basis. This study also provides recommenda- tions for future studies.
... In researching the 1987 stock market crash, author Shiller (2022) shows that investors were motivated by each other's conduct rather than by negative economic news. In more recent research, the authors Diaconaşu et al. (2022), Schaub (2022), and Wen et al. (2022) studied whether investors react to each other's behavior or information entering the digital currency market. According to Diaconaşu et al. (2022), the Bitcoin market matures with time as investor behavior fits with the uncertain information hypothesis of positive and negative events. ...
... According to Schaub (2022), Bitcoin and Ethereum overreact to such situations, but Tether has a significant value reversal following atypically positive events. Wen et al. (2022) provide evidence of intraday return predictability in the digital currency market. Other actively traded digital currencies, such as Ethereum, Litecoin, and Ripple, show evidence of intraday timing. ...
The present research focuses on the phenomenon of cryptocurrency market overreactions, especially examining the behavior of Bitcoin, DASH, EOS, Ethereum, Lisk, Litecoin, Monero, NEO, Quantum, Ripple, Stellar, and Zcash from January 2, 2018, to March 1, 2023. The findings show that there are both positive and negative autocorrelations, which might result in lowered volatility and more moderate fluctuations in prices. These results possess the potential to assist investors in making well-informed choices since they are less susceptible to being influenced by exaggerated reactions to news or information hitting the market. However, before investing in cryptocurrency markets, investors should exercise caution and carefully examine their risk tolerance, since market circumstances may change quickly, making it impossible to perform consistently profitable trades.
... Correlation between volume and frequency is found to be inversely proportional . Thus, future research topics include detecting price manipulation in cryptocurrencies at intraday level, using feature selection techniques to get improved accuracy, analysing the impact of hard and soft forks on intraday reversal and momentum, assessing time-based momentum strategy during price crashes and sharp recovery by using high-frequency data (Wen et al. 2022), and examining casual patterns between Bitcoin and other cryptocurrencies. Additionally, the volume-return relationship of cryptocurrencies (Balcilar et al. 2017; can be examined with the inclusion of seasonality and or volatility effects. ...
... The contention put forth by the author is that the surplus volatility observed in the market could have potentially engendered an unwarranted sense of assurance among domestic investors, thereby inducing them to engage in speculative conduct. In recent times, Diaconaşu et al. (2022), Schaub (2022) and Wen et al. (2022) conducted research to examine whether investors respond to one another's behaviors or to information that emerges in the digital currency market. Diaconaşu et al. (2022) conducted an analysis of the maturity of the Bitcoin market and examined the response of investors to unexpected favorable and unfavorable information. ...
... The revaluation patterns of the three cryptocurrencies following atypical positive events exhibits mixed evidence. The study conducted by Wen et al. (2022) poses findings that suggest the presence of predictability in intraday return within the cryptocurrency market. The predictability is observed to encompass both intraday dynamics and reversal. ...
This chapter aims to analyze the price efficiency of Bitcoin (BTC), DASH, EOS, Ethereum (ETH), LISK, Litecoin (LTC), Monero, NEO, QUANTUM, RIPPLE, STELLAR, and ZCASH in their weak form between March 1, 2018 and March 1, 2023 and determine whether they experience overreactions. The results show that cryptocurrencies exhibit positive and negative autocorrelations, which can reduce volatility and moderate price fluctuations. The results also show persistence in cryptocurrency returns, suggesting long-term trends or market patterns that individual and institutional investors can exploit. It is essential to recognize that cryptocurrencies are characterized by a high degree of complexity and instability. Investors need to monitor market trends and make the necessary adjustments to their investment strategies to anticipate market changes.
... Investors can find profit-making strategies by implementing trend-following. For example, investors can profit from ongoing price movement by implementing momentum strategies (Wen, Bouri, Xu, & Zhao, 2022). This involves buying cryptocurrencies that exhibit positive price momentum and selling assets that have negative price momentum. ...
... In our current contribution, we want to use a cutting-edge methodology to generate price forecasts for cryptocurrencies along with common stocks, the predictability of which is becoming one of the core research questions involving cryptocurrencies as a specific financial asset. Indeed, there is mixed evidence in the literature about the profitability and value of market sentiment for prediction and trading purposes [23][24][25][26][27][28]. More generally, because there is broad evidence that machine learning methods are efficient for prediction tasks and are explicitly designed to approximate complex nonlinear dependencies, for asset pricing problems [29] we adopt an artificial neural network (ANN) model. ...
Cryptocurrencies are nowadays seen as an investment opportunity, since they show some peculiar features, such as high volatility and diversification properties, that are triggering research interest into investigating their differences with traditional assets. In our paper, we address the problem of predictability of cryptocurrency and stock trends by using data from social online communities and platforms to assess their contribution in terms of predictive power. We extend recent developments in the field by exploiting a combination of stochastic neural networks (NNs), an extension of standard NNs, natural language processing (NLP) to extract sentiment from Twitter, and an external evolutionary algorithm for optimal parameter setting to predict the short-term trend direction. Our results point to good and robust accuracy over time and across different market regimes. Furthermore, we propose to exploit recent advances in sentiment analysis to reassess its role in financial forecasting; in this way, we contribute to the empirical literature by showing that predictions based on sentiment analysis are not found to be significantly different from predictions based on historical data. Nonetheless, compared to stock markets, we find that the accuracy of trend predictions with sentiment analysis is on average much higher for cryptocurrencies.
... In 2020, keywords related to behavioral finance, such as "sentiment", became prevalent (Anamika et al. 2021;Jha et al. 2021;Feldman and Liu 2023;Almeida 2023), which shows a research interest in the way that human behavior affects the decision-making processes of investors. More recently, by 2022, trending topics included expressions such as "bitcoin", (Dias et al. 2022;Wang et al. 2022;Wen et al. 2022;Liu et al. 2022;Benlagha and Hemrit 2023) and "artificial neural networks" (Şengüler and Mehmet 2022; Nametala et al. 2023), reflecting the trending research topics and suggesting potential areas for future investigation. ...
This article aims to contribute to the academic knowledge in the field of scientific production regarding decision support tools for investments in the capital market, specifically focusing on fundamental analysis, technical analysis, and Ichimoku dynamics. Bibliometric analysis, following the three main laws (Bradford’s Law, Lotka’s Law, and Zipf’s Law), was employed to evaluate scientific production, identify publication patterns, and uncover gaps and collaboration networks over the last thirty years. To achieve these objectives, 1710 relevant academic publications on the topic were analyzed and retrieved from the Web of Science (WOS) database, pertaining to the last 30 years, between 1990 and 22 May 2023. The significance of this article lies in the contributions of the findings, which advance scientific knowledge by identifying gaps in the knowledge and research, particularly in the limited literature on Ichimoku; our review reveals a growing trend of research in this area. Another notable conclusion is the emergence of new research topics and areas of interest, as well as the identification of collaboration networks among authors, institutions, and countries. Moreover, the article provides valuable insights for financial professionals and investors who are interested in applying these methodologies as methods for price forecasting. The highlighted results support investment decision making, trading strategies, and portfolio management.
