Quantile Regression under Semi-Competing Risks Data with Continuous Covariates
碩士 === 國立中正大學 === 數理統計研究所 === 100 === This thesis focuses on the quantile regression analysis for semi-competing risks data with continuous covariates. Since the non-terminal event may be dependently censored by the terminal event, the estimation of quantile regression parameters becomes difficult....
Main Authors: | , |
---|---|
Other Authors: | |
Format: | Others |
Language: | en_US |
Published: |
2012
|
Online Access: | http://ndltd.ncl.edu.tw/handle/02060267344630650343 |
id |
ndltd-TW-100CCU00477002 |
---|---|
record_format |
oai_dc |
spelling |
ndltd-TW-100CCU004770022015-10-13T21:01:52Z http://ndltd.ncl.edu.tw/handle/02060267344630650343 Quantile Regression under Semi-Competing Risks Data with Continuous Covariates 連續型解釋變數下半競爭風險資料之分量迴歸 Meng-Feng Tsai 蔡孟峰 碩士 國立中正大學 數理統計研究所 100 This thesis focuses on the quantile regression analysis for semi-competing risks data with continuous covariates. Since the non-terminal event may be dependently censored by the terminal event, the estimation of quantile regression parameters becomes difficult. In order to make inference on the non-terminal event, we have to make some extra assumptions. Thus, we consider the copula function to specify the dependence between the non-terminal event time and the terminal event time. We adopt the Kernel Smoothing technique to estimate the coefficients of quantile regression for semi-competing risks data with continuous covariates. For the bandwidth selection of Kernel Smoothing technique, we propose two cross-validation methods to handle it. According to the simulation studies, the performances of our proposed approach display well and we examine the asymptotic behavior by graphs. We also apply our approach into the Bone Marrow Transplant data for illustration. Jin-Jian Hsieh 謝進見 2012 學位論文 ; thesis 61 en_US |
collection |
NDLTD |
language |
en_US |
format |
Others
|
sources |
NDLTD |
description |
碩士 === 國立中正大學 === 數理統計研究所 === 100 === This thesis focuses on the quantile regression analysis for semi-competing risks data with continuous covariates. Since the non-terminal event may be dependently censored by the terminal event, the estimation of quantile regression parameters becomes difficult. In order to make inference on the non-terminal event, we have to make some extra assumptions. Thus, we consider the copula function to specify the dependence between the non-terminal event time and the terminal event time.
We adopt the Kernel Smoothing technique to estimate the coefficients of quantile regression for semi-competing risks data with continuous covariates. For the bandwidth selection of Kernel Smoothing technique, we propose two cross-validation methods to handle it. According to the simulation studies, the performances of our proposed approach display well and we examine the asymptotic behavior by graphs. We also apply our approach into the Bone Marrow Transplant data for illustration.
|
author2 |
Jin-Jian Hsieh |
author_facet |
Jin-Jian Hsieh Meng-Feng Tsai 蔡孟峰 |
author |
Meng-Feng Tsai 蔡孟峰 |
spellingShingle |
Meng-Feng Tsai 蔡孟峰 Quantile Regression under Semi-Competing Risks Data with Continuous Covariates |
author_sort |
Meng-Feng Tsai |
title |
Quantile Regression under Semi-Competing Risks Data with Continuous Covariates |
title_short |
Quantile Regression under Semi-Competing Risks Data with Continuous Covariates |
title_full |
Quantile Regression under Semi-Competing Risks Data with Continuous Covariates |
title_fullStr |
Quantile Regression under Semi-Competing Risks Data with Continuous Covariates |
title_full_unstemmed |
Quantile Regression under Semi-Competing Risks Data with Continuous Covariates |
title_sort |
quantile regression under semi-competing risks data with continuous covariates |
publishDate |
2012 |
url |
http://ndltd.ncl.edu.tw/handle/02060267344630650343 |
work_keys_str_mv |
AT mengfengtsai quantileregressionundersemicompetingrisksdatawithcontinuouscovariates AT càimèngfēng quantileregressionundersemicompetingrisksdatawithcontinuouscovariates AT mengfengtsai liánxùxíngjiěshìbiànshùxiàbànjìngzhēngfēngxiǎnzīliàozhīfēnliànghuíguī AT càimèngfēng liánxùxíngjiěshìbiànshùxiàbànjìngzhēngfēngxiǎnzīliàozhīfēnliànghuíguī |
_version_ |
1718053198378303488 |