Early Wallet Behavior and Transparency in Meme Coin Markets
Authors
Koturic, Ilija ; Baran, Magdalena Maria ; Kéri, Laura
Term
4. semester
Publication year
2025
Abstract
Denne afhandling undersøger adfærd på markeder for meme-mønter ved at kombinere analyse af on-chain transaktioner med Reddit-sentiment og emnemodellering (topic modelling). Fokus er tre Solana-tokens—$TRUMP, $MELANIA og $LIBRA—på Solana-blokkæden. On-chain analyse betyder, at man undersøger det offentlige register over transaktioner. Wallets (blokkædeadresser, der opbevarer og handler tokens) analyseres for mønstre, der kan pege på manipulation. En "mistankescore" udvikles for at markere wallets, hvis adfærd ligner manipulerende aktivitet. Resultaterne viser, at fortjenesterne er stærkt koncentreret hos en lille gruppe tidlige eller på anden måde begunstigede wallets, mens de fleste deltagere får økonomiske tab. Reddit-data indikerer, at følelser som hype, FOMO (frygten for at gå glip af noget) og social identitet får folk til at fortsætte med at investere, også efter at den profitable fase er forbi. Afhandlingen vurderer også blokkædetransparens. Selvom dataene teknisk set er offentlige, begrænses mulighederne for analyse af dårlig tilgængelighed og økonomiske barrierer. Projektet udvikler desuden visuelle værktøjer, der giver et mere fortolkeligt overblik over meme-mønter. Samlet set tyder fundene på, at ulighed og manipulation er almindelige i disse markeder.
This thesis examines behavior in meme coin markets by combining on-chain transaction analysis with Reddit sentiment and topic modeling. It focuses on three Solana tokens—$TRUMP, $MELANIA, and $LIBRA—on the Solana blockchain. On-chain analysis means studying the public record of transactions. Wallets (blockchain addresses that hold and trade tokens) are analyzed to detect patterns that may indicate manipulation. A "suspicion score" is introduced to flag wallets whose activity resembles manipulative behavior. Results show that profits are concentrated among a small group of early or otherwise advantaged wallets, while most participants incur losses. Reddit data indicates that emotions such as hype, FOMO (fear of missing out), and social identity encourage continued buying even after the profitable phase has passed. The thesis also evaluates blockchain transparency. Although the data is technically public, effective analysis is limited by poor accessibility and financial barriers. To help, the project creates visual tools that present meme coin activity in a more interpretable way. Overall, the findings suggest that inequality and manipulation are common features of these markets.
[This summary has been rewritten with the help of AI based on the project's original abstract]
Documents