Machine Learning-based Multi-Asset Portfolio Optimization: Evidence from Stocks, Funds and Bonds

DOI

10.2991/978-94-6239-801-6_144How to use a DOI?

Keywords

Machine Learning; Multi-asset; Portfolio Optimization

Abstract

With the background of significant fluctuations in the global financial market, traditional single investment strategy has demonstrated characteristics such as high volatility, and poor risk diversification. To enhance the stability of investment strategies, increase the rate of return, and address the existing issues of current strategies, this paper will conduct research on diversified investment portfolios through machine learning methods. This article mainly selects six types of financial assets: stocks, government bonds, gold, crude oil, consumer funds, and money market wealth management. By calculating the monthly rate of return, first, using the correlation analysis to get the diversified value of different assets. Then, the assets are classified into low, medium and high-risk categories based on their returns and fluctuations through the K-means algorithm. Meanwhile, using linear regression model is used to predict the stability and future trends of various assets. According to the analysis results, the diversified asset investment model constructed will be compared with a single stock investment model. The experimental results prove that a diversified investment portfolio has lower volatility, and a higher Sharpe ratio, which can effectively diversify risks and enhance the stability of investment returns. This study demonstrates that machine learning can assist in the allocation and screening of investment assets.

Copyright

© 2026 The Author(s)

Open Access

Open Access This chapter is licensed under the terms of the Creative Commons Attribution-NonCommercial 4.0 International License (http://creativecommons.org/licenses/by-nc/4.0/), which permits any noncommercial use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license and indicate if changes were made.

Download article (PDF)

TY  - CONF
AU  - Runzhe Yin
PY  - 2026
DA  - 2026/10/23
TI  - Machine Learning-based Multi-Asset Portfolio Optimization: Evidence from Stocks, Funds and Bonds
BT  - Proceedings of the 2026 2nd International Conference on Financial Innovation and Marketing Management (FIMM 2026)
PB  - Atlantis Press
SP  - 1385
EP  - 1395
SN  - 2352-5428
UR  - https://doi.org/10.2991/978-94-6239-801-6_144
DO  - 10.2991/978-94-6239-801-6_144
ID  - Yin2026
ER  -