Document Detail


Sparse and stable Markowitz portfolios.
MedLine Citation:
PMID:  19617537     Owner:  NLM     Status:  MEDLINE    
Abstract/OtherAbstract:
We consider the problem of portfolio selection within the classical Markowitz mean-variance framework, reformulated as a constrained least-squares regression problem. We propose to add to the objective function a penalty proportional to the sum of the absolute values of the portfolio weights. This penalty regularizes (stabilizes) the optimization problem, encourages sparse portfolios (i.e., portfolios with only few active positions), and allows accounting for transaction costs. Our approach recovers as special cases the no-short-positions portfolios, but does allow for short positions in limited number. We implement this methodology on two benchmark data sets constructed by Fama and French. Using only a modest amount of training data, we construct portfolios whose out-of-sample performance, as measured by Sharpe ratio, is consistently and significantly better than that of the naïve evenly weighted portfolio.
Authors:
Joshua Brodie; Ingrid Daubechies; Christine De Mol; Domenico Giannone; Ignace Loris
Related Documents :
1492347 - Organ transplants, foreign nationals, and the free rider problem.
11909657 - Evaluation of scientific evidence using bayesian networks.
2137967 - Short self-retaining trocar sleeves for laparoscopic surgery.
4009827 - Microspike approximator for vasovasostomy.
9582777 - Evaluation of an hiv/aids continuing education program.
4053197 - Environmental problems of runners.
Publication Detail:
Type:  Journal Article; Research Support, Non-U.S. Gov't; Research Support, U.S. Gov't, Non-P.H.S.     Date:  2009-07-15
Journal Detail:
Title:  Proceedings of the National Academy of Sciences of the United States of America     Volume:  106     ISSN:  1091-6490     ISO Abbreviation:  Proc. Natl. Acad. Sci. U.S.A.     Publication Date:  2009 Jul 
Date Detail:
Created Date:  2009-08-11     Completed Date:  2009-09-28     Revised Date:  2010-01-29    
Medline Journal Info:
Nlm Unique ID:  7505876     Medline TA:  Proc Natl Acad Sci U S A     Country:  United States    
Other Details:
Languages:  eng     Pagination:  12267-72     Citation Subset:  IM    
Affiliation:
Program in Applied and Computational Mathematics, Princeton University, Princeton, NJ 08544-1000, USA.
Export Citation:
APA/MLA Format     Download EndNote     Download BibTex
MeSH Terms
Descriptor/Qualifier:
Algorithms*
Computer Simulation*
Industry / standards
Models, Theoretical
Reproducibility of Results

From MEDLINE®/PubMed®, a database of the U.S. National Library of Medicine


Previous Document:  Gliomas induce and exploit microglial MT1-MMP expression for tumor expansion.
Next Document:  Why kinesin is so processive.