Skip to content
All library documents

Sliding-Window Estimation for Time-Varying Log-Optimal Portfolios

Article arXiv papers · Author: Pei-Ting Wang et al.
Read the original

Summary

The document describes a data-driven sliding-window method for solving a log-optimal portfolio problem. Rather than assigning one fixed set of portfolio weights, the approach produces weights that vary over time. This offers a way to adapt portfolio allocations using a moving window of data, although the document does not explain the window length, the estimation procedure, or how often weights are updated.

The authors report empirical studies in which the proposed approach achieves a higher cumulative rate of return than the classical log-optimal portfolio. The evidence is stated in broad terms: no markets, sample periods, numerical results, or risk-adjusted comparisons are provided. As a result, the reported return advantage alone does not establish how the method performs after transaction costs or across different settings. The document also does not specify constraints or other practical details needed to reproduce the strategy.

Key ideas

  • The approach uses a sliding window to solve a log-optimal portfolio problem.
  • It produces time-varying weights instead of fixed portfolio allocations.
  • Empirical studies report higher cumulative returns than the classical log-optimal portfolio.
  • The document does not state the tested markets, sample periods, or numerical results.
  • It provides no details on transaction costs, portfolio constraints, or implementation.

Tags

Portfolio constructionStatisticsBacktesting
Full text
# On Data-Driven Log-Optimal Portfolio: A Sliding Window Approach


# On Data-Driven Log-Optimal Portfolio: A Sliding Window Approach









In this paper, we propose a data-driven sliding window approach to solve a log-optimal portfolio problem. In contrast to many of the existing papers, this approach leads to a trading strategy with time-varying portfolio weights rather than fixed constant weights. We show, by conducting various empirical studies, that the approach possesses a superior trading performance to the classical log-optimal portfolio in the sense of having a higher cumulative rate of returns.

Shown in full with attribution under the source's licence. Licence: abstract CC0

This summary was written by Stratmill's research agent from the original; it is not a copy of the source.