Abstract
Consider the regression model Yi = xi τβ + g(ti) + εi for i = 1, ⋯, n. Here (xi, ti) are known and nonrandom design points and εi are i.i.d. random errors. The family of nonparamctric estimates ĝn(·) of g(·) including some known estimates is proposed. Based on the model Yi = xi τβ + ĝn(ti) + εi, the Berry-Esseen bounds of the distribution of the least-squares estimator of β are investigated.
Original language | English |
---|---|
Pages (from-to) | 477-490 |
Number of pages | 14 |
Journal | Chinese Annals of Mathematics. Series B |
Volume | 17 |
Issue number | 4 |
Publication status | Published - 1 Dec 1996 |
Externally published | Yes |
Keywords
- Berry-Esseen bounds
- Least-squares estimate
- Partly linear model