Hello, Login

Member Login

Forgot Password?

Join Us

to start. Not a member? Join Today!
LinkedIn Join us on
Investment Management Information
“Bridging the theory & practice of investment management”
Email
Advanced Search →
  • Home
  • Journal
    • About
    • Subscribe to the Journal
      • Subscriptions
      • Library Subscriptions
    • Harry M. Markowitz Award
    • Submit a Paper
      • Article Guidelines
      • Practitioner’s Guidelines
    • Reprints & Permissions
    • Advertising
  • Conferences
    • JOIM Conference Events
    • About
    • Membership
    • Board Members
    • Sponsorship
  • Library Access
  • Contact
  • Help

0 comments / 23/02/2017 / dbailey / Archives, Articles

Stock Portfolio Design and Backtest Overfitting

Volume 15, No. 1, 2017
David H. Bailey, Jonathan M. Borwein and Marcos López de Prado

In mathematical finance, backtest overfitting connotes the usage of historical market data to develop an investment strategy, where too many variations of the strategy are tried, relative to the amount of data available. Backtest overfitting is now thought to be a primary reason why investment models and strategies that look good on paper often disappoint in practice. Models and strategies suffering from overfitting typically target the specific idiosyncrasies of a limited dataset, rather than any general behavior, and, as a result, often perform erratically when presented with new data.

In this study, we address overfitting in the context of designing a mutual fund or investment portfolio as a weighted collection of stocks. Very often a newly minted equity-based fund of this type has been designed by an exhaustive computer-based search of some sort to obtain an optimal weighting that exhibits excellent performance based, say, on the past 10 or 20 years’ historical market data, and the fund often highlights this backtest performance.

In the present paper, we illustrate why this back test-driven portfolio design process often fails to deliver real-world performance. We have developed a computer program that, given any desired performance profile, designs a portfolio consisting of common securities, such as the constituents of the S&P 500 index, that achieves the desired profile based on in sample back test data. We then show that these portfolios typically perform erratically on more recent, out-of-sample data. This is symptomatic of statistical overfitting. Less erratic results can be obtained by restricting the portfolio to only positive-weight components, but then the results are quite unlike the target profile on both in-sample and out-of-sample data.

0 comments… add one

Cancel reply

Leave a Comment

Next Article: The Revolving Door

Previous Article: Picking “Winners” Funds

JOIM

    Library Access

    Subscribe to the Journal
    Submit a Paper
    Harry M. Markowitz Award
    Editorial Board
    Upcoming Conferences

Member Login

Forgot Password?
Join Us

Search Journal

  • Archives
  • Articles
  • Book Reviews
  • Case Studies
  • Insight
  • Practitioner's Digest
  • Special Issues
  • Surveys and Crossovers
  • ailmanen
  • Alexander Healy
  • Alistair Sinclair
  • Allan Eberhart
  • alo@mit.edu
  • Amir E. Khandani
  • Andrea Reed
  • Andrei Morozov
  • Andrew Kalotay
  • Andrew W. Lo
  • Anthony Bova
  • Anthony Tessitore
  • Antony Ghee
  • Anureet Saxena
  • Arik Ben Dor
  • Atanu Saha
  • Avner Wolf
  • Bac Van Luu
  • Ben Meng
  • Bernd Hanke
  • Bing Han
  • bodiez
  • Book Review
  • Bradford Cornell
  • Bradley Jones
  • Brian Granger
  • Brian Hayes
  • Brian Hurst
  • Brian J. Henderson
  • Burton G. Malkiel
  • BXL4po
  • cakici
  • Carr Bettis
  • Charles E. Harris
  • Charles Hyde
  • Cheol Eun
  • Constantin Petrov
  • Courtney Edwards
  • Craig French
  • D. Blake
  • Dale Gray
  • Daniel Bradley
  • Daniel Garcia
  • David Esch
  • David Ikenberry
  • David Tien
  • dbailey
  • DEAM
  • Dean Leistikow
  • Deepak Agrawal
  • Denis Chaves
  • Dennis Capozza
  • Eleanor Xiaoqing Xu
  • Emanuel Derman
  • englenyu
  • Eric Fielding
  • Eric Leeper
  • Ethan Namvar
  • Eugene Y. Lee
  • Evan Gatev
  • Fayez Elayan
  • Felix Goltz
  • fjones
  • gathanassakos
  • George Cejnek
  • George Chacko
  • ggarvey
  • Gilles Hilary
  • Gwendolyn Webb
  • Haim Levy
  • Harold Bierman
  • Harry Kat
  • HarryHMM
  • Hayne Leland
  • Hersh Shefrin
  • hguirguis
  • Hoje Jo
  • Honghui Chen
  • Hyuna Park
  • Indrajit Mitra
  • Iuliana Ismailescu
  • J. Peters Williamson
  • Jack Treynor
  • JackBogle
  • Jacob Boudoukh
  • Jacob Sisk
  • James Batterman
  • James Bennett
  • James Doran
  • Jamie Lee
  • Jasmina Hasanhodzic
  • Jason Fink
  • Jason Greene
  • Jason Hsu
  • Jason Scott
  • Javier Estrada
  • jhull
  • Jitka Hilliard
  • Jochen Andritzky
  • John Campbell
  • John H. Boyd
  • John Liechty
  • John M. Mulvey
  • John McConnell
  • Jose Menchero
  • Joseph Cerniglia
  • Joshua Livnat
  • jrbohn
  • jxiong
  • Jyh-Huei Lee
  • Kalok Chan
  • Kathryn Barraclough
  • Kay Giesecke
  • kcbrown
  • kosband
  • kothari
  • lando
  • Lauren Cohen
  • Lawrence Harris
  • lebaron
  • lgoldberg
  • Li Cai
  • Licheng Sun
  • Lionel Martellini
  • Louis Chan
  • Luca Barone
  • Malcolm Baker
  • Manuel Ammann
  • Marcin Kacperczyk
  • Mark B. Wise
  • Mark Coppejans
  • Mark Griffiths
  • Mark Latham
  • Mark Seasholes
  • Mark Wolfson
  • martinfridson
  • Michael Cliff
  • Michael Cooper
  • Michael Dempsey
  • Michael Stein
  • Michael Stutzer
  • Miikka Tauren
  • Mila Getmansky
  • Milind Sharma
  • Minder Cheng
  • Mitch Towner
  • mkritzman
  • mlevy
  • mmorey
  • mshimizu
  • mstatman
  • Murali Jagannathan
  • Navneet Arora
  • Nawalkha
  • Nicole Boyson
  • Noel Amenc
  • Ozgur Akay
  • panagora
  • Pankaj K. Jain
  • Patrick Augustin
  • Patrick de Fontnouvelle
  • Paul Hanouna
  • Paul Koch
  • Paul Kupiec
  • Paul Lajbcygier
  • Peter A. Lee
  • Peter Fledelius
  • Pranay Gupta
  • ptufano
  • Renato Staub
  • RGClarke
  • Riccardo Rebonato
  • Richard Bookstaber
  • Richard Roll
  • rmichaud
  • Robert Arnott
  • Robert B. Mendelson
  • Robert Hendershott
  • Robert Jarrow
  • Robert Michaud
  • Robert S. Harris
  • Robert Snigaroff
  • Roger
  • Ronald T. Wilcox
  • Ronnie Sadka
  • Ross Miller
  • Sandra M. Forman
  • Sang Bin Lee
  • Sarah Takaki
  • Scott Thomas
  • Seoyoung Kim
  • Sharon Garyn-Tal
  • Sheridan Titman
  • Sherry Jarrell
  • shill
  • Simon Myrgren
  • srdas
  • Stephen Brown
  • Stephen Lawrence
  • stumpp
  • Sugata Ray
  • Suresh Govindaraj
  • Terrence Hendershott
  • the JOIM
  • Thomas OBrien
  • Timo Korkeamaki
  • tom.ho
  • Tony Elavia
  • tony.kao
  • tphilips
  • Travis Sapp
  • Turan Bali
  • vbhansali
  • Vikas Agarwal
  • Vitaly Serbin
  • Wayne E. Ferson
  • Wayne H. Wagner
  • Wei Chen
  • Wei Huang
  • William Fung
  • Xi Li
  • Xiaofeng Shi
  • Xiaojing Dong
  • Xiaojing Dong
  • Yong Chen
  • Zhehan Ong
  • Zhirong Chen

    Article Categories

    • Archives
    • Articles
    • Book Reviews
    • Case Studies
    • Insight
    • Practitioner's Digest
    • Special Issues
    • Surveys and Crossovers

    Recent Comments

      JOIM

        About the JOIM
      • Library Access
      • Subscribe to the Journal
      • Submit a Paper
      • Editorial Board
      • Harry M. Markowitz Award
      • Reprints & Permissions
      • Advertising
      • Industry Links
      • Terms and Conditions

      JOIM Conference Series

      • About
      • Upcoming Conferences
      • Membership
      • Board Members
      • Sponsorship Opportunities
      • Terms & Conditions
      Speaker Reimbursement Policy

      Contact

      Journal Of Investment Management (JOIM)
      3658 Mt. Diablo Blvd., Suite 200
      Lafayette, CA 94549
      www.joim.com

      customerservice @ joim.com
      (925) 299-7800

      Copyright 2019 — Journal Of Investment Management design by SEO Web Designers