Show simple item record

dc.contributor.authorCarnegie, Nicole B.
dc.contributor.authorWu, James
dc.date.accessioned2020-11-12T18:25:19Z
dc.date.available2020-11-12T18:25:19Z
dc.date.issued2019-09
dc.identifier.citationCarnegie, Nicole Bohme, and James Wu. “Variable Selection and Parameter Tuning for BART Modeling in the Fragile Families Challenge.” Socius: Sociological Research for a Dynamic World 5 (January 2019): 237802311982588. doi:10.1177/2378023119825886.en_US
dc.identifier.issn2378-0231
dc.identifier.urihttps://scholarworks.montana.edu/xmlui/handle/1/16018
dc.description.abstractOur goal for the Fragile Families Challenge was to develop a hands-off approach that could be applied in many settings to identify relationships that theory-based models might miss. Data processing was our first and most time-consuming task, particularly handling missing values. Our second task was to reduce the number of variables for modeling, and we compared several techniques for variable selection: least absolute selection and shrinkage operator, regression with a horseshoe prior, Bayesian generalized linear models, and Bayesian additive regression trees (BART). We found minimal differences in final performance based on the choice of variable selection method. We proceeded with BART for modeling because it requires minimal assumptions and permits great flexibility in fitting surfaces and based on previous success using BART in black-box modeling competitions. In addition, BART allows for probabilistic statements about the predictions and other inferences, which is an advantage over most machine learning algorithms. A drawback to BART, however, is that it is often difficult to identify or characterize individual predictors that have strong influences on the outcome variable.en_US
dc.description.sponsorshipEunice Kennedy Shriver National Institute of Child Health and Human Development (R01HD36916R01HD40421R01HD39135) | Russell Sage Foundationen_US
dc.language.isoen_USen_US
dc.rights© This manuscript version is made available under the CC-BY-NC 4.0 licenseen_US
dc.rights.urihttp://creativecommons.org/licenses/by-nc/4.0/en_US
dc.titleVariable Selection and Parameter Tuning for BART Modeling in the Fragile Families Challengeen_US
dc.typeArticleen_US
mus.citation.journaltitleSociusen_US
mus.citation.volume5en_US
mus.relation.collegeCollege of Letters & Scienceen_US
mus.relation.departmentMathematical Sciences.en_US
mus.relation.universityMontana State University - Bozemanen_US
mus.data.thumbpage4en_US


Files in this item

Thumbnail

This item appears in the following Collection(s)

Show simple item record

© This manuscript version is made available under the CC-BY-NC 4.0 license
Except where otherwise noted, this item's license is described as © This manuscript version is made available under the CC-BY-NC 4.0 license