Essays on the impact of social media on cryptocurrency returns: Cross-platform analysis

Dai, Yanchen 2025. Essays on the impact of social media on cryptocurrency returns: Cross-platform analysis. PhD Thesis, Cardiff University. Item availability restricted.

Preview	PDF - Accepted Post-Print Version Download (7MB) | Preview
	PDF (Cardiff University Electronic Publication Form) - Supplemental Material Restricted to Repository staff only Download (112kB)

Abstract

This thesis examines the impact of investor attention, sentiment, and influencer accounts on cryptocurrency returns across multiple digital platforms. Given that cryptocurrency markets are characterised by speculative dynamics, limited fundamental valuation, and pronounced investor irrationality, behavioural biases significantly affect asset pricing. Using comprehensive daily data from nine major cryptocurrencies (2019–2023) and rigorous quantitative methodologies—including Autoregressive Distributed Lag–Error Correction (ARDL-ECM) and Time-Varying Parameter Vector Autoregression (TVP-VAR) models—this study systematically analyses cross-platform interactions involving social media (X, Reddit, YouTube, TikTok) and search-based platforms (Google, Wikipedia). The findings demonstrate that investor attention significantly shapes cryptocurrency returns, with search-based metrics offering strong long-term predictive capability, whereas social media attention predominantly drives short-term speculative volatility. Further, investor sentiment on X displays notable asymmetry, as positive sentiment consistently exerts stronger market impacts than negative sentiment, highlighting the distinct speculative nature of cryptocurrency trading. Additionally, the thesis identifies clear attention spillover patterns transitioning from initial social media engagement to subsequent structured search activities, aligning with theoretical expectations of investor information-seeking behaviour. Sentiment spillovers exhibit cyclical dynamics, systematically moving from optimism to pessimism, amplifying market volatility. The study also reveals significant short-term market reactions driven by influencer accounts, particularly affecting meme tokens, with personal authenticity and neutral messaging having substantial, sustained impacts. Overall, this thesis provides robust empirical evidence of how digital platforms and influencer-account characteristics collectively shape cryptocurrency market dynamics, enhancing the understanding of behavioural drivers underlying cryptocurrency valuation and volatility.

Item Type:	Thesis (PhD)
Date Type:	Completion
Status:	Unpublished
Schools:	Schools > Business (Including Economics)
Date of First Compliant Deposit:	8 August 2025
Last Modified:	08 Aug 2025 13:51
URI:	https://orca.cardiff.ac.uk/id/eprint/180308

Actions (repository staff only)

Edit Item