Abstract
In this paper we propose a cross-validation selection criterion to determine asymptotically the correct model among the family of all possible partially linear models when the underlying model is a partially linear model. We establish the asymptotic consistency of the criterion. In addition, the criterion is illustrated using two real sets of data.
Original language | English |
---|---|
Pages (from-to) | 754-772 |
Number of pages | 19 |
Journal | Journal of Complexity |
Volume | 17 |
Issue number | 4 |
DOIs | |
Publication status | Published - 1 Jan 2001 |
Keywords
- Dimensional reduction
- Linear regression
- Model selection
- Nonlinear regression
- Nonlinear time series
- Nonparametric regression
- Semiparametric regression