An importance and frequency-based approach for sparse online portfolios
Published:
Mar 31, 2025
Volume:
23
Keywords:
Portfolio optimization
Online Gradient Descent
Empirical Mode Decomposition
Risk measures
Brazilian stock market
Abstract
Portfolio optimization has advanced since Markowitz, yet dynamically adjusting weights remains a challenge. This study applies Online Gradient Descent (OGD) and Empirical Mode Decomposition (EMD) to construct and optimize portfolios in the Brazilian stock market. By decomposing financial time series into fast, medium, and slow modes, we build correlation-based networks and select stocks using centrality measures. Portfolios (5–30 stocks) are optimized via OGD and evaluated against a benchmark. Results show that small RIS portfolios consistently outperform large RIS portfolios in CAGR and VaR. The fast mode performs best, with all fast small RIS portfolios surpassing the benchmark. The mixed mode (γ = 0.2, 0.35) achieves superior results, with the best portfolios exceeding an 18% CAGR. Overall, 36% of mixed mode portfolios outperform the benchmark, with small RIS portfolios dominating the optimal region. These findings demonstrate the effectiveness of EMD-based stock selection and OGD for risk-adjusted returns.
How to cite
Gustavo Ribeiro de Oliveira Roque, Carlos Cristiano Hasenclever Borges, João Daniel Madureira Yamim. An importance and frequency-based approach for sparse online portfolios. Brazilian Review of Finance, v. 23, n. 1, 2025. p. e202505. DOI: 10.12660/rbfin.v23n1.2025.93034.
References
Acerbi, C. and Tasche, D. (2002). Expected shortfall: A natural coherent alternative to value at risk, Economic Notes, 31(2), 379–388. https://doi.org/10.1111/1468-0300.00091
Algoet, P.H. and Cover, T.M. (1988). Asymptotic optimality and asymptotic equipartition properties of log-optimum investment, The Annals of Probability, 876–898. https://www.jstor.org/stable/2243845
Bell, R. and Cover, T.M. (1988). Game-theoretic optimal portfolios, Management Science, 34(6), 724–733. https://doi.org/10.1287/mnsc.34.6.724
Bell, R.M. and Cover, T.M. (1980). Competitive optimality of logarithmic investment, Mathematics of Operations Research, 5(2), 161–166. https://doi.org/10.1287/moor.5.2.161
Cover, T.M. (1991). Universal portfolios, Mathematical Finance, 1(1), 1–29. https://doi.org/10.1111/j.1467-9965.1991.tb00002.x
Huang, N.E. et al. (1998). The empirical mode decomposition and the Hilbert spectrum for nonlinear and non-stationary time series analysis, Proceedings of the Royal Society of London. Series A: Mathematical, Physical and Engineering Sciences, 454(1971), 903–995. https://doi.org/10.1098/rspa.1998.0193
Kelly, J.L. (1956). A new interpretation of information rate, The Bell System Technical Journal, 35(4), 917–926. https://doi.org/10.1002/j.1538-7305.1956.tb03809.x
Mantegna, R.N. (1999). Hierarchical structure in financial markets, The European Physical Journal B-Condensed Matter and Complex Systems, 11, 193–197. https://doi.org/10.1007/s100510050929
Markowitz, H. (1952). The utility of wealth, Journal of Political Economy, 60(2), 151–158. https://doi.org/10.1086/257177
Pozzi, F., Di Matteo, T. and Aste, T. (2013). Spread of risk across financial markets: Better to invest in the peripheries, Scientific Reports, 3(1), 1665. https://doi.org/10.1038/srep01665
Tumminello, M. et al. (2005). A tool for filtering information in complex systems, Proceedings of the National Academy of Sciences, 102(30), 10421–10426. https://doi.org/10.1073/pnas.0500298102
Yang, L., Zhao, L. and Wang, C. (2019). Portfolio optimization based on empirical mode decomposition, Physica A: Statistical Mechanics and its Applications, 531, 121813. https://doi.org/10.1016/j.physa.2019.121813
Zinkevich, M. (2003). Online convex programming and generalized infinitesimal gradient ascent, Proceedings of the 20th International Conference on Machine Learning (ICML-03). https://cdn.aaai.org/ICML/2003/ICML03-120.pdf