Year: 2025
Type: article
Abstract: The following research develops methodologies of portfolio construction focused on liquidity and risk-adjusted return maximization. It considered a dataset comprising 40 stocks from the S&P 500, S&P 400, and S&P 600 indices for five years, 2019-2024, and compared three different weighting schemes: filtered risk-adjusted weighting, equal weighting, and non-filtered risk-adjusted weighting. Each strategy is compared by looking at annualized return, volatility, Sharpe ratio, and transaction cost. The methodologies used include filtering based on liquidity, calculating weight inversely proportional to the volatility, and implementing an equal-weighted allocation. As opposed to the nonfiltered strategy, which uses all stocks to maximize diversification, filtered and equal-weighted methodologies will be targeting subsets of very specific stocks. The results indicate that the non-filtered strategy has the highest annualized return, 6.30%, and Sharpe ratio, 4.817, among the three methods, although it has higher transaction costs than the other two. On the other hand, the filtered strategy, though cost-efficient, underperforms due to limited diversification. Equal weighting is simple but not optimized for risk-adjusted returns. This research emphasizes how paramount liquidity considerations are in portfolio strategies. Perhaps future research might develop dynamic allocation models based on real-time data and machine learning techniques. Such a development would increase portfolio performance, while keeping a balance of cost efficiency with diversification.
DOI: 10.54097/ra1xqk40
Author: Minghe Liu
Institution: University College London
Language: English
Subfield: Finance
Domain: Social Sciences
Open Access status: diamond
OpenAlex ID: W4413406697