Abstract
A regression model with a nonnegativity constraint on the dependent variable, known as censored median regression model, is considered. Under some mild conditions, the LAD estimate of the regression coefficient is shown to be strongly consistent. Furthermore, its convergence rate and Bahadur strong representation are also obtained.
Original language | English (US) |
---|---|
Pages (from-to) | 155-168 |
Number of pages | 14 |
Journal | Science in China, Series A: Mathematics |
Volume | 48 |
Issue number | 2 |
DOIs | |
State | Published - Feb 2005 |
All Science Journal Classification (ASJC) codes
- Mathematics(all)
Keywords
- Bahadur representation
- Censored regression
- Convergence rate
- LAD estimator
- Strong consistency