分位点

Quantile regression models are important statistical models, which are more powerful than least square regression and least absolute deviation regression in statistical analysis, and can supply much more attractive inference results for the datasets.

分位点回归模型是一种重要的统计模型，它比最小二乘回归和LAD回归具有更强的统计分析能力，能给数据集提供更加丰富的统计推断结果。

We discuss asymptotically best linear unbiased estimators of the Logistic population based on selected order statistics and give the formula of computing the ABLUE, the variance and the covariance of the ABLUE in limit, and properties of the ABLUE, and so on. We give the optimum chosen of spacing which is the ABLUE with maximum asymptotic relative efficiency based on complete sample when selected order statistics number is less than 10, and obtain its maximum asymptotic relative efficiency.

讨论基于Logistic总体Ⅱ型截尾样本的若干个样本分位数的总体分布参数的近似最佳线性无偏估计；给出了该近似最佳线性无偏估计的计算公式，估计方差与协方差的极限表达式及估计量的大样本性质等；给出了全样本，样本分位点个数固定，但不超过10情形下，使上述近似最佳线性无偏估计有最大相对联合估计效率时，样本分位数选取法及最大相对联合估计效率；为大样本情形下，该近似最佳线性无偏估计的应用提供了理论依据。

This paper uses mixture density networks to forecast the time-varying conditional density values of financial time series, and puts forth a new numerical algorithm to calculate Expected Shortfall. The application in Hong Kong Hangseng index approves that Value-at-Risk at high probability levels is accurate, and that it also has good forecasting ability in fluctuation. MDNs can effectively describe the empirical distribution of returns and the volatility mechanism, and is a new model for measuring financial market risk.

基于混合密度网络模型估计金融时间序列的时变条件密度，提出采用数值模拟方法计算Expected Shortfall的新途径，对香港恒生指数的实证研究表明，VaR的预测精度在高端分位点较好，模型评估指标反映出预测效果良好，混合密度网络可以有效地描述收益的经验分布统计特征和波动规律，是一种有效的金融市场风险测量模型。

The right and left of the box are at the third and first quartiles. Therefore, the length of the box equals the interquartile range, and the box itself represents the middle 50 percent of the observations.

图框左右分别位于第3和第1四分位点，图框的总长度正好等于一个四分位距，代表了中间观测数据的50%。

After giving a detailed review of the related research of this field in the second chapter, we developed our own method : nonparametric quantile regression,in the third chapter.

本文结构安排如下：第一章对所研究的问题进行了描述；第二章对以往研究进行了详细总结之后，我们在第三章提出来本文的研究方法：非参数分位点回归法，并介绍了此方法。

It is defined as the strength figure above which 95 % of the population of all possible strength measurements of the concrete in question may be expected to lie (95 % fractile).

对于强度数值超过混凝土所有可能强度测量总量的95%，则数据可能是错误的（95%分位点）。

We used generalized estimating equations to estimate odds ratios for the presence of CAC across quintiles of linolenic acid.

我们用无显著特点的评价方程评价了相对于亚麻酸的五分位点的冠状动脉中钙化的动脉粥样硬化斑块存在的比率。

Furthermore, out of 497 fAFLP markers, 80 special bands were found to be able to distinguish the four groups from each other and may be applied for germplasm characterization and molecular assistant classification of Meretrix clam.4 Molecular classification of two species of Meretrix clam based on fAFLP and ITS sequences4.1 The results of fAFLP maker analysis of S, G and W showed that each group had their own specific loci among which there were 53 special loci in W group, much more than those of S group (14) and G group (21). Among the 53 loci, nine were all dominant loci. These unique loci could be taken as molecular markers to distinguish W from other groups. The genetic similarity indexes and distance matrix between S and G groups were 0.9585 and 0.0424 respectively, but the genetic similarity indexes and distance matrix between W group and S or G group was 0.7939 or 0.7941, and 0.2308 or 0.2305 respectively. The results revealed that significant difference existed between W and S or G groups in molecular genetic structure. The phylogenetic trees by the methods of UPGMA and NJ also indicated that S and G populations were very closely related, while W population was a relatively independent cluster, lying beyond the species which S and G belong to.4.2 The internal transcribed spacer region of the rDNA from S group, G group and W group were PCR amplified and sequenced. The results showed that the size of ITS ranged between 1266-1269bp in W group, while those in G and S groups were 1614bp and 1520bp respectively. The GC content ranged 62.32-62.62% in W group while it was 61.77% in G group. The genetic distances between three populations of W group were 0.001～0.003, but it was 0.110 or 0.147 respectively between W group and G group or S group. Phylogenetic trees by NJ method also showed that G group was very closely related to S group, while W group was a relatively independent cluster.

在457个总扩增位点中找出了53个W的特有位点，远多于S群体(14)和G(21)群体，而且在53个特有位点中有9个出现频率为100%的位点，这些位点可以作为区分其它2个群体的特征性标记；S– G群体特有的位点有112个，其中有4个位点出现频率为100%，可作为S– G群体区别于W群体的特征性标记。S群体和G群体间的遗传相似性系数为0.9585，遗传距离只有0.0424，在NJ和UPGMA法构建的亲缘关系的树状图上均首先聚在一起，说明二者的亲缘关系很近，应属于种内群体间的关系；而W与S和G的遗传相似性系数均较小(0.7939和0.7941)，相对遗传距离很大而且十分相近(0.2308和0.2305)，在亲缘关系树状图上单独分出一支，也表明W与S和G群体间的亲缘关系较远。4.2 ITS序列比较分析通过对白壳文蛤、山东文蛤和广西文蛤的ITS序列扩增电泳、PCR-RFLP分析和ITS序列分析发现，W的ITS序列长度在1266-1269 bp，而S和与G的ITS序列总长度分别为1520 bp和1614 bp；从ITS1和ITS2长度来看，W分别为739-741 bp和316-317 bp，S为895 bp和414 bp，G为987 bp和416 bp；而从ITS碱基组成来看，W的GC含量在62.32-62.62%之间，而G群体为61.77%。W的3个壳色不同群体间的遗传距离仅0.001、0.002和0.003，S与G群体间的遗传距离是0.010，说明W群体内变异很小，而S与G群体间已出现明显的遗传分化，但还均属于种内群体间的遗传变异；而W与G和S的遗传距离分别达到0.110、0.147，两个类群差异显著，已远超出种内群体间的遗传变异。

In this study, We typed the the gene of β〓-AR from genomic DNA of asthmatic patients and made the further analysis: The distributing condition of genetic polymorphisms at positions 16 and 27 in β〓 AR coding region and at position 19 in 5'Leader Cistron (5'LC) among southwest Chinese asthmatic population; The relationship between the each polymorphisms at the above three positions and the asthmatic phenotypes (FEV〓 or nocturnal asthma); The effect of each polymorphisms at the above three positions on bronchodilator responsiveness induced by β〓- agonists(initial bronchodilator response, pharmaco-dynamic); The relationship between the haplotypes of polymorphisms in the 5'LC or in positions 16, 27 of β〓AR coding block and asthmatic phenotypes or bronchodilator responsiveness induced by β〓-agonists.

实验通过对哮喘人群的β〓-AR基因进行分型，探讨：β〓-AR编码区16、27位点及5'前导肽(5'LeaderCistron，5'LC)19位点遗传多态性在中国西南地区哮喘人群中的分布情况；三个位点多态性与哮喘临床表现型的关系；三个位点多态性对β〓受体激动剂反应性(支气管舒张反应，舒张反应动力学)的影响；16、27位点及5'前导肽19位点遗传多态性单倍型与哮喘临床表现型及β〓受体激动剂反应性的关系。

After amplying a 2.2kb fragment form the PPV-SC1 RF-DNA,we clone the fragment into pMD 18-T,named pTNSl.The whole sequence which is 1989 bp long was determined by sequencing, including the complete ORF of PPV-SC1 NS1 which encoding 662 amino acids.Alignment of pairs of sequence indicates that there are 98% and 99% similary with other porcine parvovirus strains Kresse and NADL-2, respectively. Multiple sequence alignment discloses that there are a few difference between ppv-scl nsl gene and other ppv nsl gene: A-G at 39nt,T-C at 153nt,A-G at 175nt, A-C at 1117nt, A-C at 1535nt .Alternative codon in ppv-scl nsl have distinctly different frequentfy by codonbias analysis at EMBOSS(http://genopole.toulouse.inra.fr/bioinfo/emboss). Thereis not distinct hydrophobicity and transmenbrane helices in ppv-scl nsl protein. Struction domain anslysis of PPV-SC1 NS1 protein indicate that there are a ATP/GTP-binding site motif A at 398-405,16 Protein kinase C phosphorylation site,21 Casein kinase II phosphorylation site,and 3 cAMP/cGMP-dependent protein kinase phosphorylation site.At the same time ,there is a same motif between ppv-scl nsl and Poxvirus D5 protein-like which may share in the same fuction which is necessary during virion duplication.

将PPV-SC1 NS1序列与其他PPV NS1基因进行多序列比对，结果显示，PPV-SC1 NS1与其他的PPV NS1的同源性较高，仅存在个别的差异，分别是第39位A→G，第153位T→C，第175位A→G，第1117位A→C，第1535位A→C；同源搜索比较表明，PPV-SC1与PPV NS1同源性可达98％、99％，与其他的细小病毒NS1基因也存在很大的保守性；密码子偏向性分析结果表明PPV-SC1 NS1基因在同一氨基酸的不同密码子的选择上存在一定的偏向性；PPV-SC1 NS1蛋白总体上说具有亲水性不存在明显的疏水性区段，用swiss TMPRED软件预测PPV-SC1 NS1的跨膜区，返回的结果并没有得到有显著意义的跨膜区的存在；根据基于motif数据库的结构域预测，PPV-SC1 NS1的第393-415位氨基酸残基存在潜在的ATP／GTP结合位点，该蛋白还存在16个蛋白激酶C磷酸化位点，21个酪蛋白激酶2磷酸化位点，3个cAMP-／cGMP依赖蛋白激酶磷酸化位点，PPV-SC1 NS1蛋白与POX_D5(痘病毒D5蛋白)具有一致的保守结构域，推测NS1可能与POX_D5有类似的功能。

The split between the two groups can hardly be papered over.

这两个团体间的分歧难以掩饰。

This approach not only encourages a greater number of responses, but minimizes the likelihood of stale groupthink.

这种做法不仅鼓励了更多的反应，而且减少跟风的可能性。