Leveraging sentiment connectedness for portfolio optimisation: a minimum connectedness approach

PurposeThis study proposes a novel portfolio optimisation approach that leverages minimum online sentiment connectedness. We investigate the effectiveness of this portfolio optimisation technique.Design/methodology/approachThe minimum online sentiment connectedness portfolio utilises sentiment analysis to construct portfolios that minimise sentiment spillovers, measured by the pairwise directional connectedness index. We compare the minimum negative news sentiment connectedness portfolios to traditional methods like the minimum variance portfolio (MVP) and the equally weighted portfolio (EWP).FindingsOur findings demonstrate that minimum sentiment connectedness portfolios outperform conventional approaches in terms of cumulative returns, while the MVP outperforms in risk-adjusted returns (Sharpe ratio). Furthermore, we employ the concept of hedging effectiveness to show that minimum negative news connectedness portfolios significantly reduce the risk of the overall portfolio.Originality/valueThis research contributes to the growing body of knowledge on sentiment analysis in portfolio management, suggesting its potential to improve investment decision-making.