Variable Screening Based on Combining Quantile Regression

• Author / Creator

  Shi, Qian

• This thesis develops an efficient quantile-adaptive framework for linear and nonlinear variable screening with high-dimensional heterogeneous data.
  Inspired by the success of various variable screening methods, especially in the quantile-adaptive framework, we develop a more efficient variable screening procedure. Both the classical linear regression model and the nonlinear regression model are investigated. In the thesis, the information over different quantile levels are combined, which can be implemented in two ways. The first one is the (weighted) average quantile estimator based on a (weighted) average of quantile regression estimators at single quantiles. The other one is the (weighted) composite quantile regression estimator based on a (weighted) quantile loss function.
  Simulation studies are conducted to investigate the fine performance of the finite sample. A real data example is also analyzed.

• Subjects / Keywords

  • Average quantile regression
  • B-spline approximations
  • Variable screening
  • Quantile regression
  • Composite quantile regression

• Graduation date

  Fall 2014

• Type of Item

  Thesis

• Degree

  Master of Science

• DOI

  https://doi.org/10.7939/R3416T718

• License

  This thesis is made available by the University of Alberta Libraries with permission of the copyright owner solely for non-commercial purposes. This thesis, or any portion thereof, may not otherwise be copied or reproduced without the written consent of the copyright owner, except to the extent permitted by Canadian copyright law.