Universal portfolios via context trees

Suleyman S. Kozat, Andrew Carl Singer, Andrew J. Bean

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

In this paper, we consider the sequential portfolio investment problem considered by Cover [3] and extend the results of [3] to the class of piecewise constant rebalanced portfolios that are tuned to the underlying sequence of price relatives. Here, the piecewise constant models are used to partition the space of past price relative vectors where we assign a different constant rebalanced portfolio to each region independently. We then extend these results where we compete against a doubly exponential number of piecewise constant portfolios that are represented by a context tree. We use the context tree to achieve the wealth of a portfolio selection algorithm that can choose both its partitioning of the space of the past price relatives and its constant rebalanced portfolio within each region of the partition, based on observing the entire sequence of price relatives in advance, uniformly, for every bounded deterministic sequence of price relative vectors. This performance is achieved with a portfolio algorithm whose complexity is only linear in the depth of the context tree per investment period. We demonstrate that the resulting portfolio algorithm achieves significant gains on historical stock pairs over the algorithm of [3] and the best constant rebalanced portfolio.

Original languageEnglish (US)
Title of host publication2008 IEEE International Conference on Acoustics, Speech and Signal Processing, ICASSP
Pages2093-2096
Number of pages4
DOIs
StatePublished - Sep 16 2008
Event2008 IEEE International Conference on Acoustics, Speech and Signal Processing, ICASSP - Las Vegas, NV, United States
Duration: Mar 31 2008Apr 4 2008

Publication series

NameICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings
ISSN (Print)1520-6149

Other

Other2008 IEEE International Conference on Acoustics, Speech and Signal Processing, ICASSP
CountryUnited States
CityLas Vegas, NV
Period3/31/084/4/08

Fingerprint

partitions

Keywords

  • Context tree
  • Investment
  • Piecewise models
  • Portfolio
  • Universal

ASJC Scopus subject areas

  • Electrical and Electronic Engineering
  • Signal Processing
  • Acoustics and Ultrasonics

Cite this

Kozat, S. S., Singer, A. C., & Bean, A. J. (2008). Universal portfolios via context trees. In 2008 IEEE International Conference on Acoustics, Speech and Signal Processing, ICASSP (pp. 2093-2096). [4518054] (ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings). https://doi.org/10.1109/ICASSP.2008.4518054

Universal portfolios via context trees. / Kozat, Suleyman S.; Singer, Andrew Carl; Bean, Andrew J.

2008 IEEE International Conference on Acoustics, Speech and Signal Processing, ICASSP. 2008. p. 2093-2096 4518054 (ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings).

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Kozat, SS, Singer, AC & Bean, AJ 2008, Universal portfolios via context trees. in 2008 IEEE International Conference on Acoustics, Speech and Signal Processing, ICASSP., 4518054, ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings, pp. 2093-2096, 2008 IEEE International Conference on Acoustics, Speech and Signal Processing, ICASSP, Las Vegas, NV, United States, 3/31/08. https://doi.org/10.1109/ICASSP.2008.4518054
Kozat SS, Singer AC, Bean AJ. Universal portfolios via context trees. In 2008 IEEE International Conference on Acoustics, Speech and Signal Processing, ICASSP. 2008. p. 2093-2096. 4518054. (ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings). https://doi.org/10.1109/ICASSP.2008.4518054
Kozat, Suleyman S. ; Singer, Andrew Carl ; Bean, Andrew J. / Universal portfolios via context trees. 2008 IEEE International Conference on Acoustics, Speech and Signal Processing, ICASSP. 2008. pp. 2093-2096 (ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings).
@inproceedings{b108ddf5cfa9446daea9b6addc95eee6,
title = "Universal portfolios via context trees",
abstract = "In this paper, we consider the sequential portfolio investment problem considered by Cover [3] and extend the results of [3] to the class of piecewise constant rebalanced portfolios that are tuned to the underlying sequence of price relatives. Here, the piecewise constant models are used to partition the space of past price relative vectors where we assign a different constant rebalanced portfolio to each region independently. We then extend these results where we compete against a doubly exponential number of piecewise constant portfolios that are represented by a context tree. We use the context tree to achieve the wealth of a portfolio selection algorithm that can choose both its partitioning of the space of the past price relatives and its constant rebalanced portfolio within each region of the partition, based on observing the entire sequence of price relatives in advance, uniformly, for every bounded deterministic sequence of price relative vectors. This performance is achieved with a portfolio algorithm whose complexity is only linear in the depth of the context tree per investment period. We demonstrate that the resulting portfolio algorithm achieves significant gains on historical stock pairs over the algorithm of [3] and the best constant rebalanced portfolio.",
keywords = "Context tree, Investment, Piecewise models, Portfolio, Universal",
author = "Kozat, {Suleyman S.} and Singer, {Andrew Carl} and Bean, {Andrew J.}",
year = "2008",
month = "9",
day = "16",
doi = "10.1109/ICASSP.2008.4518054",
language = "English (US)",
isbn = "1424414849",
series = "ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings",
pages = "2093--2096",
booktitle = "2008 IEEE International Conference on Acoustics, Speech and Signal Processing, ICASSP",

}

TY - GEN

T1 - Universal portfolios via context trees

AU - Kozat, Suleyman S.

AU - Singer, Andrew Carl

AU - Bean, Andrew J.

PY - 2008/9/16

Y1 - 2008/9/16

N2 - In this paper, we consider the sequential portfolio investment problem considered by Cover [3] and extend the results of [3] to the class of piecewise constant rebalanced portfolios that are tuned to the underlying sequence of price relatives. Here, the piecewise constant models are used to partition the space of past price relative vectors where we assign a different constant rebalanced portfolio to each region independently. We then extend these results where we compete against a doubly exponential number of piecewise constant portfolios that are represented by a context tree. We use the context tree to achieve the wealth of a portfolio selection algorithm that can choose both its partitioning of the space of the past price relatives and its constant rebalanced portfolio within each region of the partition, based on observing the entire sequence of price relatives in advance, uniformly, for every bounded deterministic sequence of price relative vectors. This performance is achieved with a portfolio algorithm whose complexity is only linear in the depth of the context tree per investment period. We demonstrate that the resulting portfolio algorithm achieves significant gains on historical stock pairs over the algorithm of [3] and the best constant rebalanced portfolio.

AB - In this paper, we consider the sequential portfolio investment problem considered by Cover [3] and extend the results of [3] to the class of piecewise constant rebalanced portfolios that are tuned to the underlying sequence of price relatives. Here, the piecewise constant models are used to partition the space of past price relative vectors where we assign a different constant rebalanced portfolio to each region independently. We then extend these results where we compete against a doubly exponential number of piecewise constant portfolios that are represented by a context tree. We use the context tree to achieve the wealth of a portfolio selection algorithm that can choose both its partitioning of the space of the past price relatives and its constant rebalanced portfolio within each region of the partition, based on observing the entire sequence of price relatives in advance, uniformly, for every bounded deterministic sequence of price relative vectors. This performance is achieved with a portfolio algorithm whose complexity is only linear in the depth of the context tree per investment period. We demonstrate that the resulting portfolio algorithm achieves significant gains on historical stock pairs over the algorithm of [3] and the best constant rebalanced portfolio.

KW - Context tree

KW - Investment

KW - Piecewise models

KW - Portfolio

KW - Universal

UR - http://www.scopus.com/inward/record.url?scp=51449083099&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=51449083099&partnerID=8YFLogxK

U2 - 10.1109/ICASSP.2008.4518054

DO - 10.1109/ICASSP.2008.4518054

M3 - Conference contribution

AN - SCOPUS:51449083099

SN - 1424414849

SN - 9781424414840

T3 - ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings

SP - 2093

EP - 2096

BT - 2008 IEEE International Conference on Acoustics, Speech and Signal Processing, ICASSP

ER -