Positive graphical lasso estimation of sparse inverse covariance matrices

Graduate Project

Public Deposited

Citeable URL: https://ir.library.oregonstate.edu/concern/graduate_projects/2227mr28q

Descriptions

Attribute Name	Values
Creator	Logan, Paul
Abstract	We explore the possibility of estimating sparse inverse covariance matrices when for scientific reasons the covariance matrix is restricted to be a non-negative matrix. The process mirrors the graphical lasso process developed by Friedman and others(2008) that did not have this additional constraint.Accordingly, the Lasso procedure is done through coordinate descent. To easily add the constraint, we modified the LARS function created by Efron and others(2004) to perform positive Lasso (pLasso) estimation. The process is demonstrated on several time series generated datasets to clearly show the effectiveness and limitations.
Resource Type	Capstone Project
Date Available	2017-05-01T21:19:37+00:00
Date Issued	2017-05-01
Degree Level	Master's
Degree Name	Master of Science (M.S.)
Degree Field	Statistics
Degree Grantor	Oregon State University
Commencement Year	2016
Advisor	Mondal, Debashis
Committee Member	Emerson, Sarah Wickham, Charlotte
Academic Affiliation	Statistics
Rights Statement	In Copyright
Publisher	Oregon State University
Peer Reviewed	No
Language	English [eng]
Replaces	http://hdl.handle.net/1957/60662

Thumbnail	Title	Date Uploaded	Visibility	Actions
	LoganPaulR2016.pdf	2017-07-25	Public	Download