Chance-constrained programming for Cryptocurrency portfolio optimization using Conditional Drawdown at Risk

Portfolio optimization is a widely studied problem in financial engineering literature. Its objective is to effectively distribute capital among different assets to maximize returns and minimize the risk of losing capital. Although portfolio optimization has been extensively investigated, there has been limited focus on optimizing portfolios consisting of cryptocurrencies, which are rapidly growing and emerging markets. The cryptocurrency market has demonstrated significant growth over the past two decades, offering potential profits but also presenting heightened risks compared to traditional financial markets. This situation creates challenges in constructing portfolios, necessitating the development of new and improved risk management models for cryptocurrency funds. This paper utilizes a new risk measurement approach called Conditional Drawdown at Risk (CDaR) in constructing portfolios within high-risk financial markets. Traditionally, portfolio optimization has been approached under certain conditions, considering risk and profit as decision criteria. However, recent approaches have addressed uncertainty in the decision-making process. To contribute to the advancement of scientific knowledge in this field, this paper proposes a new mathematical formulation of CDaR based on a chance-constrained programming (CCP) approach for portfolio optimization. To demonstrate the effectiveness of the proposed model, a practical empirical case study is conducted using real-world market data from 10 months focused on cryptocurrencies. The results obtained from this model can provide valuable guidance in making investment decisions in high-risk financial markets.