http://chineseinput.net/에서 pinyin(병음)방식으로 중국어를 변환할 수 있습니다.
변환된 중국어를 복사하여 사용하시면 됩니다.
Yunlu Jiang,Hong Li 한국통계학회 2014 Journal of the Korean Statistical Society Vol.43 No.4
In this paper, a penalized weighted composite quantile regression estimation procedureis proposed to estimate unknown regression parameters and autoregression coefficientsin the linear regression model with heavy-tailed autoregressive errors. Under some conditions,we show that the proposed estimator possesses the oracle properties. In addition, weintroduce an iterative algorithm to achieve the proposed optimization problem, and usea data-driven method to choose the tuning parameters. Simulation studies demonstratethat the proposed new estimation method is robust and works much better than the leastsquares based method when there are outliers in the dataset or the autoregressive errordistribution follows heavy-tailed distributions. Moreover, the proposed estimator workscomparably to the least squares based estimator when there are no outliers and the erroris normal. Finally, we apply the proposed methodology to analyze the electricity demanddataset.
Robust variable selection for the varying index coefficient models
Zou Hang,Jiang Yunlu 한국통계학회 2023 Journal of the Korean Statistical Society Vol.52 No.4
Recently, the statistical inference of the varying index coefficient model has been widely concerned. However, to the best of our knowledge, there has no existing robust variable selection method for the varying index coefficient model in the presence of outliers in the response and covariates. To overcome this difficulty, we develop a robust variable selection method for the varying index coefficient model via the exponential squared loss (ESL) function in this article. We first approximate nonparametric functions by B-spline basis functions and then apply the minorization-maximization (MM) algorithm and the Fisher scoring algorithm to calculate the proposed estimators. Under some mild conditions, the theoretical properties of the proposed estimators are established. Furthermore, we propose a data-driven procedure to select the tuning parameters. Some numerical simulations are conducted to illustrate the finite sample performance of the proposed method. Finally, the analysis of New Zealand workforce data reveals the merit of the proposed method.
Genetic dissection of leaf-related traits using 156 chromosomal segment substitution lines
Xi Liu,Linglong Liu,Yinhui Xiao,Shijia Liu,Yunlu Tian,Liangming Chen,Zhiquan Wang,Ling Jiang,Zhigang Zhao,Jianmin Wan 한국식물학회 2015 Journal of Plant Biology Vol.58 No.6
A two-line super-hybrid rice (Oryza sativa L.) variety [Liangyoupei9 (LYP9)] demonstrated superiority over its both parents, viz. elite inbred lines 93-11 and Pei-ai64S (PA64S), as well as other conventional hybrids, and had long been exploited in China. However, the genetic basis of its leaf-related traits, supposed to be an important component for yield potential, remains elusive. Here, initially a set of chromosome segment substitution lines (CSSLs) was constructed, in which the genome of Pei-ai64S has been introgressed into the background of 93-11. This set was developed by marker aided selection, based on 123 polymorphic SSR markers. The introgressed chromosomal segments presented in the 156 CSSLs covered 96.46% of Pei-ai64S genome. Afterwards, the CSSLs were deployed to assess the genetic basis of leaf size (length and width) and chlorophyll content of top three leaves across five different environments. The CSSLs showed transgressive segregation for all of the traits, and significant correlations were detected among most of the traits. A total of 27 quantitative trait loci (QTL) were identified on ten chromosomes, and three QTL cluster affecting related traits were found on chromosome 3, 6, and 8, respectively. Remarkably, two key QTLs, qALW3-1 and qALW3-2, both controlling the antepenultimate leaf width, were identified in all five environments, and their effect were further validated by CSSLs harboring the two QTL alleles. Our results indicate that developing CSSLs is a powerful tool for genetic dissection of quantitative traits. Meanwhile, the QTLs controlling leaf-related traits uncovered here provide useful information for marker-assisted selection in improving the performance of leaf morphology and photosynthetic ability.