Identification of Partially Linear Structure in Additive Models with an Application to Gene Expression Prediction from Sequences
Article first published online: 23 SEP 2011
© 2011, The International Biometric Society
Volume 68, Issue 2, pages 437–445, June 2012
How to Cite
Lian, H., Chen, X. and Yang, J.-Y. (2012), Identification of Partially Linear Structure in Additive Models with an Application to Gene Expression Prediction from Sequences. Biometrics, 68: 437–445. doi: 10.1111/j.1541-0420.2011.01672.x
- Issue published online: 26 JUN 2012
- Article first published online: 23 SEP 2011
- Received February 2011. Revised July 2011., Accepted July 2011.
- Bayesian information criterion;
- Oracle property;
- Partially linear additive models;
- Structure identification
Summary The additive model is a semiparametric class of models that has become extremely popular because it is more flexible than the linear model and can be fitted to high-dimensional data when fully nonparametric models become infeasible. We consider the problem of simultaneous variable selection and parametric component identification using spline approximation aided by two smoothly clipped absolute deviation (SCAD) penalties. The advantage of our approach is that one can automatically choose between additive models, partially linear additive models and linear models, in a single estimation step. Simulation studies are used to illustrate our method, and we also present its applications to motif regression.