Publication Type
PhD Dissertation
Version
publishedVersion
Publication Date
6-2025
Abstract
The dissertation consists of three chapters on behavioral finance, arbitrage activities, cryptocurrencies and ETFs. The first chapter examines the existence and trading behavior of triangular arbitrageurs in cryptocurrency markets. Using a novel account-level orderbook dataset from an Indian cryptocurrency exchange, we document persistent price deviations between USD- and INRquoted token pairs, and use the triangular arbitrage opportunities to identify arbitrageurs and noise traders. Our results show that arbitrageurs significantly outperform noise traders on their holding-based returns, after adjusting for market returns and trading fees. However, arbitrageurs are not immune to behavioral biases when making trading decisions. Using a composite behavioral bias index constructed by combining four different well-recognized psychological biases: extrapolation, disposition effects, lottery preferences and excessive trading, we find, surprisingly, arbitrageurs often exhibit higher levels of behavioral biases, and their returns are also more negatively impacted by the composite behavioral bias measure than those of noise traders. The results suggest that the classical rational assumption about arbitrageurs may be problematic when applied to newly emerged markets, such as cryptocurrencies, which could have important normative implications.
The second chapter studies how cryptocurrency market scandals could influence the trading strategies and behavioral biases of different groups of traders. The behavior of retail investors in cryptocurrencies is particularly sensitive to market incidents, such as the collapse of FTX, which significantly eroded market trust and triggered contagion effects. Given the youth and limited regulation of the crypto market, such events, including scams, hacks, and cybersecurity breaches, are not uncommon. We test how cybersecurity scandals impact investors' decision-making, using account-level data from a major Indian cryptocurrency exchange. Building on a prior classification of arbitrageurs and noise traders, our findings reveal that noise traders respond negatively to cybersecurity events, displaying an increased level of composite behavioral bias despite their lower sensitivity to bias in portfolio performance. This heightened overall behavioral bias level is primarily driven by elevated lottery preference, suggesting that cybersecurity incidents may amplify speculative tendencies among noise traders.
The third chapter explores speculative demands in the credit ETF market. Fund flows tend to contain essential information on investor expectations and demands. In this chapter, I focus on the ETF flows generated by the arbitrage activities of authorized participants when ETF prices deviate from their net asset values (NAVs). I find that high-yield corporate bond ETF flows have additional predictability on daily credit spread, even after controlling for credit mutual fund flows. This finding highlights the importance of retail investor sentiment and their speculative demands on the broader corporate bond market.
Degree Awarded
PhD in Business (Finance)
Discipline
Finance and Financial Management
Supervisor(s)
ZHANG, Hong
First Page
1
Last Page
140
Publisher
Singapore Management University
City or Country
Singapore
Citation
GONG, Ruxue.
Essays on empirical asset pricing. (2025). 1-140.
Available at: https://ink.library.smu.edu.sg/etd_coll/785
Copyright Owner and License
Author
Creative Commons License
This work is licensed under a Creative Commons Attribution-NonCommercial-No Derivative Works 4.0 International License.