Please use this identifier to cite or link to this item:
https://scholarbank.nus.edu.sg/handle/10635/16159
DC Field | Value | |
---|---|---|
dc.title | Subset selection in regression model | |
dc.contributor.author | HU XIAOLI | |
dc.date.accessioned | 2010-04-08T11:01:42Z | |
dc.date.available | 2010-04-08T11:01:42Z | |
dc.date.issued | 2007-06-06 | |
dc.identifier.citation | HU XIAOLI (2007-06-06). Subset selection in regression model. ScholarBank@NUS Repository. | |
dc.identifier.uri | https://scholarbank.nus.edu.sg/handle/10635/16159 | |
dc.description.abstract | For the problem of selecting subset of possible predictors, there are many kinds of criteria and methods. The LASSO (least absolute shrinkage and selection operator) approach proposed by Tibshirani (1996) is currently the most popular approach. Itsimultaneously accomplishes model estimation and variable selection.A new approach is proposed in this thesis to explore the possibility of improving LASSO. The primary motivation for developing thistechnique came from the different characteristics of discrete and continuous selection approaches. It is effective if the two are combined together, without any loss of good properties. Simulation study and real data analysis suggest that the new procedure performs better than the original LASSO in most of the cases. | |
dc.language.iso | en | |
dc.subject | LASSO, variable selection | |
dc.type | Thesis | |
dc.contributor.department | STATISTICS & APPLIED PROBABILITY | |
dc.contributor.supervisor | XIA YINGCUN | |
dc.description.degree | Master's | |
dc.description.degreeconferred | MASTER OF SCIENCE | |
dc.identifier.isiut | NOT_IN_WOS | |
Appears in Collections: | Master's Theses (Open) |
Show simple item record
Files in This Item:
File | Description | Size | Format | Access Settings | Version | |
---|---|---|---|---|---|---|
MSc Thesis of Hu Xiaoli.pdf | 639.37 kB | Adobe PDF | OPEN | None | View/Download |
Google ScholarTM
Check
Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.