We present a novel method for estimating tree-structured covariance matrices directly from observed continuous data. Specifically, we estimate a covariance matrix from observations of p continuous random variables encoding a stochastic process over a tree with p leaves. A representation of these classes of matrices as linear combinations of rank-one matrices indicating object partitions is used to formulate estimation as instances of well-studied numerical optimization problems., In particular, our estimates are based on projection, where the covariance estimate is the nearest tree-structured covariance matrix to an observed sample covariance matrix. The problem is posed as a linear or quadratic mixed-integer program (MIP) where a setting of the integer variables in the MIP specifies a set of tree topologies of the structured covariance matrix. We solve these problems to optimality using efficient and robust existing MIP solvers., We present a case study in phylogenetic analysis of gene expression and a simulation study comparing our method to distance-based tree estimating procedures.
%0 Journal Article
%1 bravo_estimating_2009
%A Bravo, Héctor Corrada
%A Wright, Stephen
%A Eng, Kevin H.
%A Keles, Sündüz
%A Wahba, Grace
%D 2009
%J Journal of Machine Learning Research
%K Integer Learning, Machine covariance expression, gene matrix, phylogenetics, programming, structure tree
%P 41--48
%T Estimating tree-structured covariance matrices via mixed-integer programming
%U http://www.ncbi.nlm.nih.gov/pmc/articles/PMC3212858/
%V 5
%X We present a novel method for estimating tree-structured covariance matrices directly from observed continuous data. Specifically, we estimate a covariance matrix from observations of p continuous random variables encoding a stochastic process over a tree with p leaves. A representation of these classes of matrices as linear combinations of rank-one matrices indicating object partitions is used to formulate estimation as instances of well-studied numerical optimization problems., In particular, our estimates are based on projection, where the covariance estimate is the nearest tree-structured covariance matrix to an observed sample covariance matrix. The problem is posed as a linear or quadratic mixed-integer program (MIP) where a setting of the integer variables in the MIP specifies a set of tree topologies of the structured covariance matrix. We solve these problems to optimality using efficient and robust existing MIP solvers., We present a case study in phylogenetic analysis of gene expression and a simulation study comparing our method to distance-based tree estimating procedures.
@article{bravo_estimating_2009,
abstract = {We present a novel method for estimating tree-structured covariance matrices directly from observed continuous data. Specifically, we estimate a covariance matrix from observations of p continuous random variables encoding a stochastic process over a tree with p leaves. A representation of these classes of matrices as linear combinations of rank-one matrices indicating object partitions is used to formulate estimation as instances of well-studied numerical optimization problems., In particular, our estimates are based on projection, where the covariance estimate is the nearest tree-structured covariance matrix to an observed sample covariance matrix. The problem is posed as a linear or quadratic mixed-integer program (MIP) where a setting of the integer variables in the MIP specifies a set of tree topologies of the structured covariance matrix. We solve these problems to optimality using efficient and robust existing MIP solvers., We present a case study in phylogenetic analysis of gene expression and a simulation study comparing our method to distance-based tree estimating procedures.},
added-at = {2017-01-09T13:57:26.000+0100},
author = {Bravo, Héctor Corrada and Wright, Stephen and Eng, Kevin H. and Keles, Sündüz and Wahba, Grace},
biburl = {https://www.bibsonomy.org/bibtex/25ecd761ae95c40833411ca65f1422a23/yourwelcome},
interhash = {43665879f7a705a37d3c62c384cf1790},
intrahash = {5ecd761ae95c40833411ca65f1422a23},
issn = {1532-4435},
journal = {Journal of Machine Learning Research},
keywords = {Integer Learning, Machine covariance expression, gene matrix, phylogenetics, programming, structure tree},
pages = {41--48},
pmcid = {PMC3212858},
pmid = {22081761},
timestamp = {2017-01-09T14:01:11.000+0100},
title = {Estimating tree-structured covariance matrices via mixed-integer programming},
url = {http://www.ncbi.nlm.nih.gov/pmc/articles/PMC3212858/},
urldate = {2012-07-28},
volume = 5,
year = 2009
}