Department of Mathematics
Transformed sufficient dimension reduction
© 2014 Biometrika Trust. We propose a general framework for dimension reduction in regression to fill the gap between linear and fully nonlinear dimension reduction. Themain idea is to first transformeach of the raw predictors monotonically and then search for a low-dimensional projection in the space defined by the transformed variables. Both user-specified and data-driven transformations are suggested. In each case, the methodology is first discussed in generality and then a representative method is proposed and evaluated by simulation. The proposed methods are applied to a real dataset.
Minimum average variance estimation, Monotone smoothing spline, Predictor transformation, Probability integral transformation, Sliced inverse regression
Source Publication Title
Oxford University Press
Link to Publisher's Edition
Wang, T., X. Guo, L. Zhu, and P. Xu. "Transformed sufficient dimension reduction." Biometrika 101.4 (2014): 815-829.