Online method adapts portfolio window size to reduce trading costs
Cost-Sensitive Online Window Size Selection for Portfolio Management
Machine Learning
Summary
Choosing how far back in time to look at stock prices matters for investment strategies, especially when markets change. This paper shows a new way to automatically pick and combine different time windows so portfolios adapt better over time without excessive trading costs. The authors treat each window choice as an expert and update their influence based on both performance and trading costs. Their approach comes with mathematical guarantees on how well it tracks the best switching strategy under realistic cost conditions.
What this means in practice
- •For quantitative finance teams: Implement adaptive portfolio strategies that minimize trading costs by dynamically selecting past data window sizes in volatile markets.
- •For algorithmic trading developers: Build online trading systems that adjust their look-back periods to changing market behavior while controlling portfolio turnover and costs.
Authors
Yi-Chen Liu, Chung-Han Hsieh
Abstract
This paper investigates cost-sensitive online window size selection for portfolio management under changing market conditions. Specifically, we propose a two-level framework that constructs portfolios using candidate window sizes and dynamically aggregates them through online learning. By treating candidate window sizes as ``experts,'' we dynamically update their aggregation weights using turnover-inclusive losses. Moreover, we derive finite-horizon cost-sensitive tracking-regret bounds that account for turnover of the aggregated portfolio, with static regret as a special case. Under bounded losses and cost rates, suitably tuned Fixed Share achieves asymptotically no tracking regret for sublinear switching budgets, with Hedge covering the static case.