2007
DOI: 10.3905/jpm.2007.690606
|View full text |Cite
|
Sign up to set email alerts
|

Shrinking the Covariance Matrix

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
1
1
1
1

Citation Types

8
49
1

Year Published

2010
2010
2016
2016

Publication Types

Select...
5
2
1

Relationship

1
7

Authors

Journals

citations
Cited by 76 publications
(58 citation statements)
references
References 10 publications
8
49
1
Order By: Relevance
“…Specifically, we plan to evaluate the out-of-sample (ex-post) performance of the tangency portfolio that is constructed using a two-block covariance matrix. Stein, C. (1955) Ledoit and Wolf (2003), Jagannathan and Ma (2003), Disatnik and Benninga (2007), and DeMiguel et al (2009).…”
Section: Discussionmentioning
confidence: 99%
See 2 more Smart Citations
“…Specifically, we plan to evaluate the out-of-sample (ex-post) performance of the tangency portfolio that is constructed using a two-block covariance matrix. Stein, C. (1955) Ledoit and Wolf (2003), Jagannathan and Ma (2003), Disatnik and Benninga (2007), and DeMiguel et al (2009).…”
Section: Discussionmentioning
confidence: 99%
“…Namely, we are dealing here with relatively small-sized within-block covariances. Thus, like the shrinkage estimators advocated by Ledoit and Wolf (2003, 2004a, 2004b, and the portfolios of estimators advocated by Jagannathan and Ma (2000), Disatnik and Benninga (2007), and Fletcher (2009), the block matrix has the appealing property of covariance elements which are shrunk compared to the typically large covariances of the traditional sample matrix. Not only are the large covariances those responsible for the extreme short positions that are obtained so often when the mean-variance theory is implemented in practice, but as Michaud (1989) states, inverting the sample matrix with its large covariance elements also amplifies the sampling error.…”
Section: Theoremmentioning
confidence: 99%
See 1 more Smart Citation
“…Some authors, like Wolf (2003, 2004) and Disatnik and Benninga (2007), however, argue that the estimation of the structured covariance matrix introduces specification errors, so propose a combination of structured and sample covariance. Wolf (2003, 2004) suggest shrinking extreme values of the sample covariance matrix toward the center.…”
Section: Portfolio Selectionmentioning
confidence: 99%
“…But from an application point of view, Disatnik and Benninga, using historical data of NYSE stocks, showed that the optimal portfolios produced using the Shrinkage estimators are not significantly better than those produced using the traditional sample covariance matrix, see Disatnik and Benninga (2007).…”
Section: Stein or Shrinkage Covariance Matrixmentioning
confidence: 99%