File Download
There are no files associated with this item.
Supplementary
-
Citations:
- Appears in Collections:
Conference Paper: Adaptive Lp estimation under a general class of error densities
| Title | Adaptive Lp estimation under a general class of error densities |
|---|---|
| Authors | |
| Issue Date | 2004 |
| Citation | The 6th World Congress of the Bernoulli Society for Mathematical Statistics and Probability and the 67th Annual Meeting of the Institute of Mathematical Statistics, Barcelona, Spain, 26-31 July 2004, Abstract no. 224 How to Cite? |
| Abstract | We consider the problem of Lp estimation of regression coef-
ficients under a general class of error densities. It is known
that both the convergence rate of the Lp estimator and the actual
mode of Lp estimation depend crucially on the choice of p
and the shape of the error density near the origin. We develop
a procedure for choosing p adaptively to yield highly accurate
Lp estimators. It makes use of a special algorithm to automatically
select the right mode of Lp estimation and the m out of n
bootstrap to consistently estimate the log mean squared error of
the Lp estimator. Our proposed adaptive Lp estimator is compared
with other adaptive and non-adaptive Lp estimators in a
simulation study, which confirms superiority of our procedure. |
| Persistent Identifier | http://hdl.handle.net/10722/110201 |
| DC Field | Value | Language |
|---|---|---|
| dc.contributor.author | Lai, PY | en_HK |
| dc.contributor.author | Lee, SMS | en_HK |
| dc.date.accessioned | 2010-09-26T01:55:35Z | - |
| dc.date.available | 2010-09-26T01:55:35Z | - |
| dc.date.issued | 2004 | en_HK |
| dc.identifier.citation | The 6th World Congress of the Bernoulli Society for Mathematical Statistics and Probability and the 67th Annual Meeting of the Institute of Mathematical Statistics, Barcelona, Spain, 26-31 July 2004, Abstract no. 224 | - |
| dc.identifier.uri | http://hdl.handle.net/10722/110201 | - |
| dc.description.abstract | We consider the problem of Lp estimation of regression coef- ficients under a general class of error densities. It is known that both the convergence rate of the Lp estimator and the actual mode of Lp estimation depend crucially on the choice of p and the shape of the error density near the origin. We develop a procedure for choosing p adaptively to yield highly accurate Lp estimators. It makes use of a special algorithm to automatically select the right mode of Lp estimation and the m out of n bootstrap to consistently estimate the log mean squared error of the Lp estimator. Our proposed adaptive Lp estimator is compared with other adaptive and non-adaptive Lp estimators in a simulation study, which confirms superiority of our procedure. | - |
| dc.language | eng | en_HK |
| dc.relation.ispartof | The 6th World Congress of the Bernoulli Society for Mathematical Statistics and Probability and the 67th Annual Meeting of the Institute of Mathematical Statistics | en_HK |
| dc.title | Adaptive Lp estimation under a general class of error densities | en_HK |
| dc.type | Conference_Paper | en_HK |
| dc.identifier.email | Lee, SMS: smslee@hkusua.hku.hk | en_HK |
| dc.identifier.authority | Lee, SMS=rp00726 | en_HK |
| dc.identifier.hkuros | 115395 | en_HK |