文献类型：期刊文献

英文题名：Combination Forecasting Reversion Strategy for Online Portfolio Selection

作者：Huang, Dingjiang[1,2];Yu, Shunchang[2];Li, Bin[3];Hoi, Steven C. H.[4];Zhou, Shuigeng[5,6]

机构：[1]East China Normal Univ, Sch Data Sci & Engn, Shanghai 200062, Peoples R China;[2]East China Univ Sci & Technol, Dept Math, Shanghai 200237, Peoples R China;[3]Wuhan Univ, Econ & Management Sch, Luojia Hill, Wuhan 430072, Hubei, Peoples R China;[4]Singapore Management Univ, Sch Informat Syst, 80 Stanford Rd, Singapore 639798, Singapore;[5]Fudan Univ, Sch Comp Sci, Shanghai 200433, Peoples R China;[6]Fudan Univ, Shanghai Key Lab Intelligent Informat Proc, Shanghai 200433, Peoples R China

年份：2018

卷号：9

期号：5

外文期刊名：ACM TRANSACTIONS ON INTELLIGENT SYSTEMS AND TECHNOLOGY

收录：;EI(收录号：20184005889366);WOS:【SSCI(收录号:WOS:000463797000010)，SCI-EXPANDED(收录号:WOS:000463797000010)】；

基金：This work was partially supported by the National Natural Science Foundation of China (11501204, 71401128, U1711262), the Natural Science Foundation of Shanghai (15ZR1408300), the Program of Science and Technology Innovation Action of Science and Technology Commission of Shanghai Municipality (STCSM) (17511105204), Academic Team Building Plan for Young Scholars from Wuhan University (WHU2016012), and Singapore Ministry of Education Academic Research Fund Tier 1 Grant (14-C220-SMU-016).

语种：英文

外文关键词：Portfolio selection; online learning; mean reversion; combination forecasting reversion; combination forecasting estimators

摘要：Machine learning and artificial intelligence techniques have been applied to construct online portfolio selection strategies recently. A popular and state-of-the-art family of strategies is to explore the reversion phenomenon through online learning algorithms and statistical prediction models. Despite gaining promising results on some benchmark datasets, these strategies often adopt a single model based on a selection criterion (e.g., breakdown point) for predicting future price. However, such model selection is often unstable and may cause unnecessarily high variability in the final estimation, leading to poor prediction performance in real datasets and thus non-optimal portfolios. To overcome the drawbacks, in this article, we propose to exploit the reversion phenomenon by using combination forecasting estimators and design a novel online portfolio selection strategy, named Combination Forecasting Reversion (CFR), which outputs optimal portfolios based on the improved reversion estimator. We further present two efficient CFR implementations based on online Newton step (ONS) and online gradient descent (OGD) algorithms, respectively, and theoretically analyze their regret bounds, which guarantee that the online CFR model performs as well as the best CFR model in hindsight. We evaluate the proposed algorithms on various real markets with extensive experiments. Empirical results show that CFR can effectively overcome the drawbacks of existing reversion strategies and achieve the state-of-the-art performance.