Bitcoin Pricing Kernels (PKs) are estimated using a novel data set from Deribit, the leading Bitcoin options exchange. The PKs, as the ratio between risk-neutral and physical density, dynamically reflect the change in investor preferences. Thus, the PKs improve the understanding of investor expectations and risk premiums in a new asset class. Bootstrap-based confidence bands are estimated in order to validate the results. Investors are heterogeneous in their risk profiles and preferences with respect to volatility and investment horizon. The empirical PKs turn out to be U-shaped for short-dated instruments and W-shaped for long-dated instruments. We find that investors are willing to pay a substantial risk premium to insure themselves against short-term price movements. The risk premium is smaller for longer-dated instruments and their traders are risk averse. The shape of the empirical PKs reveals the existence of a time-varying risk premium. The similarity between the shape of empirical PKs for Bitcoin and other markets that represent aggregate wealth shows that Bitcoin is becoming an established asset class.
We find evidence of systematic insider trading in cryptocurrency markets, where individuals use private information to buy coins prior to exchange listing announcements. Leveraging blockchain data, we identify the specific transactions and wallets (individuals) that consistently trade before announcements, ruling out alternative explanations. We estimate that insider trading occurs in 28-48% of cryptocurrency listings, yielding at least $30 million in trading profits. Unlike insider trading in stocks, scrutiny by authorities does not significantly reduce the level of insider trading, but pushes it underground via cryptocurrency concealment methods. These findings highlight the substantial challenges in policing cryptocurrency markets.
Non-Fungible Tokens (NFTs) are a relatively unexplored class of assets. Designing strategies to forecast NFT trends is an intricate task due to its extremely volatile nature. The market is largely driven by public sentiment and "hype", which in turn has a high correlation with conversations taking place on social media platforms like Twitter. Prior work done for modelling stock market data does not take into account the extent of impact certain highly influential tweets and their authors can have on the market. Building on these limitations and the nature of the NFT market, we propose a novel reach-aware temporal learning approach to make predictions for forecasting future trends in the NFT market. We perform experiments on a new dataset consisting of over 1.3 million tweets and 180 thousand NFT transactions spanning over 15 NFT collections curated by us. Our model (TA-NFT) outperforms other state-of-the-art methods by an average of 36%. Through extensive quantitative and ablative analysis, we demonstrate the ability of our approach as a practical method for predicting NFT trends.
<abstract> <p>This paper examines the evidence of herding in the revolutionary cryptocurrency market for the period from January 2017 to December 2020. The study employs quantile regression technique for investigating herd behaviour during market asymmetries of rising and falling returns, extreme market returns, high volatility, and the exogenous event of the COVID-19 pandemic. The results provide evidence of pronounced herding during the bull phase, extreme down-markets, and high volatility. These results indicate that herd hunch is prevalent in the cryptocurrency market as investors exhibit imitation while ignoring their own knowledge and beliefs. Also, the phenomenon is more vividly observed during the panic period of COVID-19.</p> </abstract>
Hisham Farag, Di Luo, Larisa Yarovaya, Damian Zięba
We examine the liquidity provision premium in cryptocurrency markets using the returns from the short reversal strategy. We show that returns from liquidity provision can be predicted using the volatility index, realized variance, risk aversion, crash risk, tail risk, and innovations of Tether liquidity. We also find that<br/>an increase in the liquidity provision premium is associated with a decline in liquidity, trading volume, and transaction count, as well as more withdrawals, higher fees, and greater impermanent loss on Uniswap.<br/>This suggests potential competition between centralized and decentralized exchanges. Further, the liquidity provision premium of stock markets in China and Japan positively predicts the premium of cryptocurrency markets (effect of a common shock), meanwhile that of stock markets in the US and Canada negatively predicts the premium of cryptocurrency markets (substitution effect).
Murat Tiniç, Ahmet Şensoy, Erdinç Akyıldırım, Shaen Corbet
Abstract In this article we investigate the influence that information asymmetry may have on future volatility, liquidity, market toxicity, and returns within cryptocurrency markets. We use the adverse‐selection component of the effective spread as a proxy for overall information asymmetry. Using order and trade data from the Bitfinex exchange, we first document statistically significant adverse‐selection costs for major cryptocurrencies. Also, our results suggest that adverse‐selection costs, on average, correspond to 10% of the estimated effective spread, indicating an economically significant impact of adverse‐selection risk on transaction costs in cryptocurrency markets. Finally, we document that adverse‐selection costs are important predictors of intraday volatility, liquidity, market toxicity, and returns.
Employing the standard event study methodology and the OLS market model to examine how the global pandemic announcement impacted cryptocurrencies, we test the null hypotheses that "the global pandemic declaration did not significantly impact the abnormal returns of the cryptocurrencies", and "during the global pandemic declaration, the cryptocurrencies did not experience any significant abnormal volatilities". The average abnormal return on t-2 was nearly minus 40 percent, which is the highest negative value during the 61-day event window. The cumulative average returns are significantly negative during the event window. The global pandemic news has significantly impacted cryptocurrencies and are more volatile during the outbreak. The study's findings will empower the investors to implement proper investment strategies during emergencies.
We examine the turn of the month effect (TOM) in cryptocurrency markets. In contrast to most calendar effect studies, we do not take for granted that the TOM period is the last trading day of the month up to the first three trading days (-1, 3), as Lakonishok and Smidt (1988) proposed in their seminal paper, but we employ an optimisation algorithm which tests several four-day intramonth periods. Our findings confirm the existence of the TOM effect because the most profitable four-day periods are those between the last days of one month and the first trading days of the next one [the (-1, 3) definition is included in these combinations]. We reach the conclusion that the existence of a TOM effect may not always lead to higher profits in comparison with a buy-and-hold (BnH) strategy, but it presents better returns to risk reward and it could be beneficial for investment strategies.
Hossein Jahanshahloo, Felix Irresberger, Andrew Urquhart
This paper explores and describes historical on-chain transaction data recorded on the Bitcoin blockchain, constructs a panel of all individual Bitcoin users, and computes their balances in the cross-section and over time. We run clustering algorithms to combine addresses that belong to the same user into wallets and we find that using wallets over addresses as the unit of analysis allows for economically meaningful interpretations of user behavior. We identify and divide wallets into user categories - miners, exchanges, services, retail wallets and receiving-only addresses - and observe varying activity levels and balances in the cross-section and over time, corresponding to their intended role in the Bitcoin network. By matching historical transactions with minute-level price data, we estimate wallets' realized financial return and find that these user-types not only exhibit different transaction patterns and balances, but also different levels of financial performance. Our paper highlights opportunities for novel empirical research that exploits Bitcoin wallet-level data on individual user characteristics.
Cryptocurrencies are widely known for their limited publicly available information, making it challenging to predict market returns. Technical analysis has emerged as an essential tool in this context, but its effectiveness in the cryptocurrency market remains an open question. Using data from nearly 3,000 cryptocurrencies at daily, weekly, and monthly horizons from 2013 to 2022, we systematically re-examine the efficacy of trend-based technical indicators in predicting cryptocurrency market returns and find that price-based signals are more effective in predicting short-term horizons, while volume-based signals are more powerful in predicting long-term horizons. Further analysis shows that machine learning techniques can significantly improve the performance of technical indicators, and technical indicators based on different information respond differently to the COVID-19 outbreak. These results provide direct evidence that volume imparts information to technical analysis independently of price.
The aim of this study is to investigate the volatility spillover connectedness between NFTs attention and financial markets. This paper firstly proposes a new direct proxy for the public’s attention in the NFT market: the non-fungible tokens attention index (NFTsAI), based on 590m news stories from the LexisNexis News & Business database and applies the historical decomposition to assess the historical variations of the NFTsAI. Then the empirical analysis is performed via a TVP-VAR volatility spillover connectedness model. The empirical results show that NFTsAI indicates NFT markets are dominated by cryptocurrency, DeFi, equity, bond, commodity, F.X. and gold markets. And NFT markets are volatility spillover receivers. In addition, NFT assets could impede financial contagion and have significant diversification benefits. Employing a panel pooled OLS regression model as a supplementary analysis and a GARCH-MIDAS model as a robustness test. This study reveals that NFTsAI has sufficient power to explain the return of NFT assets from a fixed effect perspective, and NFTsAI contains useful forecasting information for both short and long-term volatility of NFT markets, separately. The new NFTsAI and the empirical findings contain useful insights for risk-averse investors, portfolio managers, institutional investors, academics and financial policy regulators.