We gratefully acknowledge support from
the Simons Foundation and member institutions.
Full-text links:

Download:

Current browse context:

q-fin.PM

Change to browse by:

References & Citations

Bookmark

(what is this?)
CiteULike logo BibSonomy logo Mendeley logo del.icio.us logo Digg logo Reddit logo ScienceWISE logo

Quantitative Finance > Portfolio Management

Title: Index Tracking with Cardinality Constraints: A Stochastic Neural Networks Approach

Abstract: Partial (replication) index tracking is a popular passive investment strategy. It aims to replicate the performance of a given index by constructing a tracking portfolio which contains some constituents of the index. The tracking error optimisation is quadratic and NP-hard when taking the L0 constraint into account so it is usually solved by heuristic methods such as evolutionary algorithms. This paper introduces a simple, efficient and scalable connectionist model as an alternative. We propose a novel reparametrisation method and then solve the optimisation problem with stochastic neural networks. The proposed approach is examined with S&P 500 index data for more than 10 years and compared with widely used index tracking approaches such as forward and backward selection and the largest market capitalisation methods. The empirical results show our model achieves excellent performance. Compared with the benchmarked models, our model has the lowest tracking error, across a range of portfolio sizes. Meanwhile it offers comparable performance to the others on secondary criteria such as volatility, Sharpe ratio and maximum drawdown.
Comments: Accepted to AAAI 2020 (Oral). 8 pages, 4 figures. Camera-ready version
Subjects: Portfolio Management (q-fin.PM)
Cite as: arXiv:1911.05052 [q-fin.PM]
  (or arXiv:1911.05052v2 [q-fin.PM] for this version)

Submission history

From: Yongxin Yang [view email]
[v1] Tue, 12 Nov 2019 18:18:22 GMT (826kb,D)
[v2] Thu, 14 Nov 2019 13:52:45 GMT (832kb,D)

Link back to: arXiv, form interface, contact.