数据加载中.....
|
请使用永久网址来引用或连结此文件:
https://nccur.lib.nccu.edu.tw/handle/140.119/156766
|
| 题名: | Stablecoin depegging risk prediction |
| 作者: | 謝明華
Hsieh, Ming-Hua;Chen, Lee, Yi-His;Chiu, Yu-Fen |
| 贡献者: | 風管系 |
| 关键词: | Stablecoins;Depegging;Machine learning |
| 日期: | 2025-04 |
| 上传时间: | 2025-04-30 15:03:07 (UTC+8) |
| 摘要: | This study aims to identify and analyze key factors contributing to depegging risks in stablecoins, consolidating insights from the literature into four critical categories: trading price and volume, market information, sentiment, and volatility. Utilizing these insights, we develop predictive models using three machine learning algorithms—logistic regression, random forest, and XGBoost—to accurately and timely predict stablecoin depegging events. Our primary subjects are the top four stablecoins by daily trading volume: USDT, USDC, BUSD, and DAI. Diverging from previous studies that employed static depegging thresholds, we adopt a dynamic threshold adjusted for trading volume. Additionally, this study is the first to incorporate sentiment indicators from news sources alongside traditional on-chain price and volume data. Covering the empirical period from January 1, 2022, to December 31, 2023. Our findings confirm that significant fluctuations in mainstream cryptocurrencies (BTC and ETH) indeed influence stablecoin depegging. While past literature's instability measures provide early warning effects, the sentiment indicators surprisingly did not show significant early warning effects for our research subjects. The models developed enable crypto asset investors to predict the risk of stablecoin depegging promptly, facilitating informed investment decisions and reducing investment risks. |
| 關聯: | Pacific-Basin Finance Journal, Vol.90, 102640 |
| 数据类型: | article |
| DOI 連結: | https://doi.org/10.1016/j.pacfin.2024.102640 |
| DOI: | 10.1016/j.pacfin.2024.102640 |
| 显示于类别: | [風險管理與保險學系] 期刊論文
|
文件中的档案:
| 档案 |
描述 |
大小 | 格式 | 浏览次数 |
|---|
| index.html | | 0Kb | HTML | 24 | 检视/开启 |
|
在政大典藏中所有的数据项都受到原著作权保护.
|
