小麦骨干亲本及其后代田间试验设计和分析方法
|
摘要
在小麦育种中骨干亲本是改良品种的基础,没有骨干亲本的优良遗传性状,优良的小麦新品种是难以育成或发展起来的。不仅我国绝大部分小麦大面积推广品种是骨干亲本的后代,而且还由其衍生出许多具有广泛应用价值的亲本育种材料。但是由于试验材料是骨干亲本及其衍生后代组成的,这些亲本和它们的衍生系(后代)具有亲缘关系。对于大数量具有亲缘相似性特点的骨干亲本及其衍生后代材料的田间试验及其相应的分析方法值得探讨研究。
本研究以2007-2008年陕西杨凌矮秆小麦骨干亲本及其衍生后代共254份材料为研究对象,通过前期确定大数量骨干亲本及其衍生后代品系试验设计方案(具有分层结构的设计),后期利用分层混合模型来解决材料间相关性,估计品种效应,再利用多重比较、线性组间比较、育种值估计、多元统计、综合评价等分析方法,阐明不同世代的骨干亲本和后代的性状差异,深入分析骨干亲本及其衍生后代之间的性状演变趋势和亲本对后代的贡献大小。另外通过对2007-2008年陕西杨凌模拟产生的小区产量数据和2007-2008年山东泰安实测小区产量数据进行分析,筛选出相对效率较高且稳定的模型方法为模型的选择提供依据。最后以R语言环境为开发平台以R-Commander为设计模板,加载了相应的分析模块,方便了试验者和科研人员在今后针对具有相关性材料的大数量区域试验中进行分析操作。
1.在处理设计中针对骨干亲本衍生群内的亲缘相似性特点,按照分层设计思路,遵照区组内相似性及各衍生世代相对平衡原则,进行处理分层的不完全区组设计。在环境设计中采用双向田间主区区组排列、主区区组内处理拉丁方排列及设置主区中心对照、随机对照、侧对照设计。
2.线性固定模型和分层混合模型的14个性状处理效应估计标准误的合计值分别为187.58及120.89,分层混合模型处理估计误差比固定模型降低35.55%。可以看出采用重复内区组分层混合模型可有效的降低处理效应的估计误差,建议在大数量处理且数据缺失不平衡时使用分层混合模型进行处理效应的估计。通过ICC(组内相关系数)值,发现:其中小区产量、分蘖数、叶长、株高4个性状具有较高的ICC值说明这4个性状组内各个体间存在较高的相似性。从两种多重比较方法进行分组后可知,Tukey检验得到字母重叠的分组结果,它比较适用于样本量较少时进行深入比较和细致分析。Scott-Knott检验得到字母不重叠的分组结果,它在大样本时结果表示简洁且易于解释。因此建议在大样本时使用Scott-Knott检验的分组结果较好。
3.本研究对3个骨干亲本阿夫、碧玛4号、南大2419及其衍生系谱品种在2008年陕西杨凌进行了12个性状鉴定。从骨干亲本与衍生子一代12个性状平均值比较来看,骨干亲本阿夫将小区产量、分蘖数、抗条锈病、穗粒数、穗粒重、叶长、叶宽等7个性状的较高基因累积水平的遗传背景传递给阿夫衍生子一代,而在抽穗期、每穗小穗数、千粒重、穗长、株高等5个性状的基因累积水平由于其他亲本影响而得到改进。
骨干亲本碧玛4号将小区产量、分蘖数、抗条锈病、每穗小穗数、穗长、穗粒数、穗粒重、叶长、叶宽等9个性状的较高基因累计水平的遗传背景传递给碧玛4号衍生子二代,而在抽穗期、千粒重、株高等3个性状的基因累计水平由于其他亲本影响而得到改进。
骨干亲本南大2419将小区产量、分蘖数、每穗小穗数、叶宽等4个性状的较高基因累积水平的遗传背景传递给南大2419衍生子一代,而在抽穗期、抗条锈病、千粒重、穗长、穗粒数、穗粒重、叶长、株高等8个性状的基因累积水平由于其他亲本影响而得到改进。从9个性状的分析上看出,阿夫、碧玛4号、南大2419等3个骨干亲本的衍生子一代、子二代、子三代、子四代中都保持稳定,由此说明骨干亲本都将良好遗传背景稳定地传递给衍生后代。
从阿夫、碧玛4号、南大2419等3个骨干亲本的衍生后代12个性状育种值分析来看,阿夫亲本给后代的有利基因效应传递较大的是抗条锈病、高穗粒数、高穗粒重、长叶、宽叶等5个性状。其中有利基因效应传递较大的是抗条锈病、长叶。并且具有不利基因效应的每穗小穗数的育种值很小,由此说明阿夫亲本给后代带来不利基因的遗传负荷很小。碧玛4号亲本给后代的有利增效基因效应传递最大的是多分蘖数,但是也在低小区产量、感条锈病、低小穗数、短穗、低穗粒数、低穗粒重、短叶、窄叶、高秆等9个性状给后代带来不利基因效应的遗传负荷。南大2419亲本给后代的有利增效基因效应传递较大的是多小穗数、高千粒重、长穗、高穗粒重等4个性状,但是也在少分蘖数、感条锈病、短叶、高秆等4个性状给后代带来不利基因的遗传负荷。
4.采用欧氏距离类平均法对17个材料进行聚类分析。分为8个不同的类群:类群Ⅰ为碧玛4号;类群Ⅱ为阿夫;类群Ⅲ为碧玛4号中间亲本;类群Ⅳ为南大2419;类群Ⅴ为阿夫子一代、阿夫子二代、阿夫子三代、阿夫子四代;类群Ⅵ为南大2419子一代、南大2419子二代、南大2419子三代、南大2419子四代、南大2419子五代;类群Ⅶ为阿夫子五代;类群Ⅷ为碧玛4号子二代、碧玛4号子三代、碧玛4号子四代。通过对应分析可揭示每个类群的特征,最大程度的体现了骨干亲本及其各世代衍生后代间的相似性和差异性。通过因子载荷分析可将八个因子命名为:穗粒数量和重量因子、粒重因子、生育期因子、小区产量因子、叶宽因子、每穗小穗数因子、株高因子、叶长因子。其中穗粒数量和重量因子由于方差贡献最大达到14%,此因子对产量的形成起着决定作用。通过改进雷达图综合评价分析得到各样本综合得分,其中前五名的世代依次是:南大2419子五代、南大2419子四代、阿夫子三代、阿夫子四代、南大2419子二代,综合得分最低的是:碧玛4号。由综合得分中三个亲本基因型值和后代估计育种值来看:南大2419亲本在综合得分上贡献给后代的有利增效基因潜力较大;碧玛4号亲本在综合得分上贡献给后代的不利减效基因潜力较大.三个亲本对下代的综合传递能力:碧玛4号最强,南大2419其次,阿夫最弱。
5.通过对2007-08年陕西杨凌矮杆小麦100次模拟小区产量数据的原始线性模型进行有效性分析,结果表明:三个固定效应:重复、处理、重复内区组的多套模拟数据计算的F值中超过50%大于原始数据F值,认为三个固定效应的原始概率值可以接受,即原始线性模型方差分析结果是有效的,可利用模型产生的模拟数据进行分析。通过两个地点数据分析综合说明了三次重复简化广义格子设计分析(SGL)模型能更进一步控制田间试验误差,且增加重复数可以更为精确的估计处理效应,在山东实测数据实验中2次重复(SGL)模型就能达到三次重复随机完全区组分析(RB)模型的试验效率水平。进一步证明了大数量试验中重复次数为3次重复的(SGL)模型能有效提高试验效率。
6.本研究以R语言环境为开发平台以R-Commander为设计模板,遵循软件工程理念,采用模块化设计方法,从研究原有的混合模型分析程序、调整均值估计程序、多重比较方法程序和线性分组比较程序入手,构建了以鼠标、菜单操作的软件分析模块,并初步得到以下结果:(1)构建了混合模型分析模块,利用该模块可用于试验设计中的裂区设计分析,巢式设计分析等,试验者和科研人员可方便的在参数菜单中选择最大似然估计(ML)和限制性最大似然估计(REML)并在固定效应和随机效应参数中对分类因子各水平在不同层次上进行效应估计。(2)增加了调整均值的估计分析模块,方便在以后的协方差分析中对处理因子各水平效应均值进行有效的调整以及在非均衡数据情况下对均衡情况的边缘均值做简单地估计。(3)增加了多重比较分析、线性分组比较模块,并对经典的方法做了扩增(如:加入各种P值调整的多重比较方法和针对任意对比的Scheffe、Tukey线性分组比较法)丰富了试验者和科研人员对多重比较、线性分组比较方法的选择。
Founder parent is a base of improved variety in wheat breeding.A new and good wheat variety cannot be bred or developed without the integrated elite characters of Founder parent.Not only the descendant of Founder Parent were selected wheat varieties with large area as materials in most areas of china.But also It derives a lot of breeding parents who have a widely application.But because of experimental material are made up of founder parents and their progeny. And founder parent and its derivative progeny have a relative affinity.Field trial and the corresponding analytical methods for materials of founder parent and its derivative progeny with basing on the Large quantity and Affinities
similarity which are worth disscussing and researching. This study is consider that the 254 materials of founder parents and their progeny from Yangling dwarf wheat variety trials in 2007-2008.The scheme of design(the design of hierarchical structure) for the number of founder parent and its derivative progeny was determined in prophase.The strain effects and correlation of the material are being solved by using hierarchical mixed model,And significant differences in traits between founder parents and their progeny in different generations was illustrated using methods of Multiple Comparison,Linear Grouped Contrast,Breeding value estimates,Multivariate Statistics analysis and Comprehensive Evaluation.The evolutionary tendency and contribution between founder parents and their progeny has been analyzed more further.Through analyzing the plot yield data which would be partly simulated from YangLing in 2007-2008 of ShanXi Province,and partly measured from TaiAn in 2007-2008 of ShanDong Province.The more relative efficiency and stable models were selected by simulation.Finally,We set R language environment as Development platform,and set R-Commander as Design Template.We constructed a software analytical module.It convenient to analyze and manipulate in the correlation of the material and large number of reginal trials and in the future for experimentalist and research worker.
1. Based on the characteristic affinities similarity in founder parent and its derivative progeny of treatment design. According to the ideas of hierarchical design.“intrablock similarity”and“Relative balance on each derivative generation”should be basic principle during conducted incomplete block design with hierarchical treatment. Using Two-direction main area block Arrange in field, Latin square Arrange in intrablock treatment of main area and Setting center CK,random CK,side CK in main area during Environmental design.
2. The S.E.means on the estimated treatment effects for 14 traits of general linear model and hierarchical mixed model value were 187.58 and 120.89,respectively,the S.E.means on the estimated treatment effects of hierarchical mixed model value was lowered 35.55% than general linear model value. It is observed that hierarchical mixed model on grouping block within repeat could effectively reduce the errors on the estimated treatment effects.Suggestions are useing hierarchical mixed model to estimated treatment effects when it has a Large number of treatments and Data are missing and imbalancing.Through ICC(Interclass Correlation Coefficient),We find that Four traits:plot yield,tillernumber,leaf length,plant height,which has higher ICC values.It shows that individual have higher Similarity within class in 4 traits. according to grouping results by two methods on multiple comparisons.the letter was overlap in Tukey grouping test. It is delicate, Intensive to be Compared on small samplesize.the letter wasn’t overlap in Scott-Knott grouping test.It is succinct,easy to be explained on large samplesize. It is thus suggested that Using the scott-knott method can conduct better the grouping results on large samplesize.
3. Base on 3 founder parents:Funo,Bima4,Nanda2419 and its derivative progeny as research object.then their 12 traits were identified in YangLing of ShanXi province in 2008. In view of the means for 12 traits from founder parent and its first filial generations. Founder parent of Funo can transfer genetic background which have high cumulative level on alleles from 7 traits:plot yield, tillernumber, the ability of stripe rust-resistant, grain number per spike, grain weight per spike, leaf length, leaf width to Its first filial generations. But the cumulative level on alleles from 5 traits: heading stage, spikelet number per spike, 1000 grains weight, ear length, plant height are improved due to the influence of Other parents.
Founder parent of Bima4 can transfer genetic background which have high cumulative level on alleles from 9 traits:plot yield, tillernumber, the ability of stripe rust-resistant, spikelet number per spike, ear length,grain number per spike, grain weight per spike, leaf length, leaf width to Its second filial generations.But the cumulative level on alleles from 3 traits: heading stage, 1000 grains weight, plant height are improved due to the influence of Other parents. Founder parent of Nanda2419 can transfer genetic background which have high cumulative level on alleles from 4 traits:plot yield, tillernumber, spikelet number per spike, leaf width to Its first filial generations.But the cumulative level on alleles from 8 traits: heading stage, the ability of stripe rust-resistant, ear length,grain number per spike, grain weight per spike, leaf length,plant height are improved due to the influence of Other parents. By analysing the 9 traits .It has remained basically steady on 3 founder parents:Funo,Bima4,Nanda2419 and their first filial generations,second filial generations,third filial generations, fourth filial generations, fifth filial generations, It shows that the Founder parent can Steady transfer good genetic background to its derivative progeny.
In view of the breeding value for 12 traits from 3 founder parent:Funo,Bima4,Nanda2419 and its derivative progeny.Founder parent of Funo can transfer favorable genetic effect from 5 traits: the ability of stripe rust-resistant, high grain number per spike, high grain weight per spike, leaf length, leaf width to Its derivative progeny.And stripe rust-resistant, leaf length can transfer more favorable genetic effect to Its derivative progeny.breeding value which have unfavourable genetic effect is small on spikelet number per spike.It shows that Founder parent of Funo can transfer smaller genetic load which have unfavourable genetic effect to Its second filial generations. Founder parent of Bima4 can transfer most favorable and additive genetic effect from 1 traits: tillernumber to Its derivative progeny.But Founder parent of Bima4 can transfer genetic load which have unfavourable genetic effect to Its derivative progeny from 9 traits: low plot yield, Feeling stripe rust, less spikelet number per spike,short ear length,less grain number per spike, light grain weight per spike,short leaf, narrow leaf,high plant height.Founder parent of Nanda2419 can transfer most favorable and beneficiated genetic effect from 4 traits: more spikelet number per spike,high 1000 grains weight,long ear length, heavy grain weight per spike to Its derivative progeny.But Founder parent of Nanda2419 can transfer genetic load which have unfavourable genetic effect to Its derivative progeny from 4 traits: less tillernumber, Feeling stripe rust, short leaf, high plant height.
4. The 17 grouping materials were clustered by using Euclidean distance and group-average method.It suggested that the materials could be divided into 8 groups: GroupⅠis Bima4.GroupⅡis Funo.GroupⅢis inter parents of Bima4.GroupⅣis Nanda2419. GroupⅤare first filial generations, second filial generations, third filial generations, fourth filial generations, fifth filial generations of Funo. GroupⅥare first filial generations, second filial generations, third filial generations, fourth filial generations, fifth filial generations of Nanda2419. GroupⅦis fifth filial generations of Funo. GroupⅧare second filial generations, third filial generations, fourth filial generations of Bima4 respectively.through a Correspondence Analysis, announcing to the characteristic of each groups: The results maximum present Similarity and dissimilarity between Founder parents and its derivative progeny.Through the factor loading was estimated by principal component analysis. The 8 factors will be named:The factor of grain number and weight per spike; The factor of 1000 grains weight; The factor of growth period ; The factor of plot yield; The factor of leaf Width; The factor of spikelet number per spike; The factor of plant height; The factor of leaf length;We find that the factor of grain number per spike and grain weight whose contribution of variance accounted to 14%.This factor play a decisive role in yield-building.Through the comprehensive evaluation and analysis by improved Radar chart.We can get the comprehensive score for each sample. The top 5 generations followed by:the fifth filial generations of Nanda2419,the fourth filial generations of Nanda2419,the third filial generations of Funo,the fourth filial generations of Funo,the second filial generations of Nanda2419. The lowest comprehensive score is Bima4. In view of the Comprehensive scores for genotype value and EBVs from3 founder parent and its derivative progeny.Founder parent of Nanda2419 can transfer more favorable and beneficiated genetic effect to Its derivative progeny. Founder parent of Bima4 can transfer more unfavorable and reductive genetic effect to Its derivative progeny.Comprehensive transmission ability for 3 founder parents and its derivative progeny:The best is Bima4,The next is Nanda2419,The weakest is Funo.
5. Through the effectiveness analysis of the Primitive linear model which based on the 100 times of simulation data for plot yield from Yangling dwarf wheat variety trials in 2007-2008.The results show that Over 50 percent of F value which was calculated by the 100 times of simulation data for Three fixed effects: Repetition, Treatment,Block within Repetition is greater than F value which was calculated by raw data. We think original probability value from three fixed effects is quite acceptable. It is show that the result of ANOVA is effectiveIt is available to produced simulation data by the Primitive linear model, and It can be analyzed.
Through analysis the data from 2 locis, it has been confirmed that the simplified generalized grid design analysis (SGL) is better control the error from field experiment,and It Can be more precise estimateeffects of treatments by increasing the number of Repetition.Test efficiency level in (SGL)model on 2 replications as well as Random complete block designs(RB)model on 3 replications. It is proved that (SGL)model on 3 replications can highly increase test efficiency in large number of experiment.
6. This research based on R language environment platform.And taken R-Commander as the design of the template,established a module of the mixed models.Thismethod could be used in split plot design analysis and Nested design analysis in Design of Experiment.Experimentalist and Research worker can easily choose the maximum likelihood estimation(MLE)and the restricted estimation maximum likelihood (REML) in the parameter menu.And It can estimated the effects for each levels in grouping factors of different hierarchies in Fixed effects and random effects parameters.an increase of estimating adjustment mean modules.It can easily adjust means in each levels of treatment factors for Analysis of covariance in the future,and Marginal Means were simply estimated for assuming the balanced situation in unbalanced data.an increase of Multiple comparison analysis and Linear grouping contrasts modules,and It has increased some methods on the previous(for example:Added some methods of Multiple comparison analysis for P value adjustment and Scheffe、Tukey Linear grouping contrasts for arbitrary contrasts).It enrich the methods of Multiple comparison analysis and Linear grouping contrasts which were selected by Experimentalist and Research worker.
本研究以2007-2008年陕西杨凌矮秆小麦骨干亲本及其衍生后代共254份材料为研究对象,通过前期确定大数量骨干亲本及其衍生后代品系试验设计方案(具有分层结构的设计),后期利用分层混合模型来解决材料间相关性,估计品种效应,再利用多重比较、线性组间比较、育种值估计、多元统计、综合评价等分析方法,阐明不同世代的骨干亲本和后代的性状差异,深入分析骨干亲本及其衍生后代之间的性状演变趋势和亲本对后代的贡献大小。另外通过对2007-2008年陕西杨凌模拟产生的小区产量数据和2007-2008年山东泰安实测小区产量数据进行分析,筛选出相对效率较高且稳定的模型方法为模型的选择提供依据。最后以R语言环境为开发平台以R-Commander为设计模板,加载了相应的分析模块,方便了试验者和科研人员在今后针对具有相关性材料的大数量区域试验中进行分析操作。
1.在处理设计中针对骨干亲本衍生群内的亲缘相似性特点,按照分层设计思路,遵照区组内相似性及各衍生世代相对平衡原则,进行处理分层的不完全区组设计。在环境设计中采用双向田间主区区组排列、主区区组内处理拉丁方排列及设置主区中心对照、随机对照、侧对照设计。
2.线性固定模型和分层混合模型的14个性状处理效应估计标准误的合计值分别为187.58及120.89,分层混合模型处理估计误差比固定模型降低35.55%。可以看出采用重复内区组分层混合模型可有效的降低处理效应的估计误差,建议在大数量处理且数据缺失不平衡时使用分层混合模型进行处理效应的估计。通过ICC(组内相关系数)值,发现:其中小区产量、分蘖数、叶长、株高4个性状具有较高的ICC值说明这4个性状组内各个体间存在较高的相似性。从两种多重比较方法进行分组后可知,Tukey检验得到字母重叠的分组结果,它比较适用于样本量较少时进行深入比较和细致分析。Scott-Knott检验得到字母不重叠的分组结果,它在大样本时结果表示简洁且易于解释。因此建议在大样本时使用Scott-Knott检验的分组结果较好。
3.本研究对3个骨干亲本阿夫、碧玛4号、南大2419及其衍生系谱品种在2008年陕西杨凌进行了12个性状鉴定。从骨干亲本与衍生子一代12个性状平均值比较来看,骨干亲本阿夫将小区产量、分蘖数、抗条锈病、穗粒数、穗粒重、叶长、叶宽等7个性状的较高基因累积水平的遗传背景传递给阿夫衍生子一代,而在抽穗期、每穗小穗数、千粒重、穗长、株高等5个性状的基因累积水平由于其他亲本影响而得到改进。
骨干亲本碧玛4号将小区产量、分蘖数、抗条锈病、每穗小穗数、穗长、穗粒数、穗粒重、叶长、叶宽等9个性状的较高基因累计水平的遗传背景传递给碧玛4号衍生子二代,而在抽穗期、千粒重、株高等3个性状的基因累计水平由于其他亲本影响而得到改进。
骨干亲本南大2419将小区产量、分蘖数、每穗小穗数、叶宽等4个性状的较高基因累积水平的遗传背景传递给南大2419衍生子一代,而在抽穗期、抗条锈病、千粒重、穗长、穗粒数、穗粒重、叶长、株高等8个性状的基因累积水平由于其他亲本影响而得到改进。从9个性状的分析上看出,阿夫、碧玛4号、南大2419等3个骨干亲本的衍生子一代、子二代、子三代、子四代中都保持稳定,由此说明骨干亲本都将良好遗传背景稳定地传递给衍生后代。
从阿夫、碧玛4号、南大2419等3个骨干亲本的衍生后代12个性状育种值分析来看,阿夫亲本给后代的有利基因效应传递较大的是抗条锈病、高穗粒数、高穗粒重、长叶、宽叶等5个性状。其中有利基因效应传递较大的是抗条锈病、长叶。并且具有不利基因效应的每穗小穗数的育种值很小,由此说明阿夫亲本给后代带来不利基因的遗传负荷很小。碧玛4号亲本给后代的有利增效基因效应传递最大的是多分蘖数,但是也在低小区产量、感条锈病、低小穗数、短穗、低穗粒数、低穗粒重、短叶、窄叶、高秆等9个性状给后代带来不利基因效应的遗传负荷。南大2419亲本给后代的有利增效基因效应传递较大的是多小穗数、高千粒重、长穗、高穗粒重等4个性状,但是也在少分蘖数、感条锈病、短叶、高秆等4个性状给后代带来不利基因的遗传负荷。
4.采用欧氏距离类平均法对17个材料进行聚类分析。分为8个不同的类群:类群Ⅰ为碧玛4号;类群Ⅱ为阿夫;类群Ⅲ为碧玛4号中间亲本;类群Ⅳ为南大2419;类群Ⅴ为阿夫子一代、阿夫子二代、阿夫子三代、阿夫子四代;类群Ⅵ为南大2419子一代、南大2419子二代、南大2419子三代、南大2419子四代、南大2419子五代;类群Ⅶ为阿夫子五代;类群Ⅷ为碧玛4号子二代、碧玛4号子三代、碧玛4号子四代。通过对应分析可揭示每个类群的特征,最大程度的体现了骨干亲本及其各世代衍生后代间的相似性和差异性。通过因子载荷分析可将八个因子命名为:穗粒数量和重量因子、粒重因子、生育期因子、小区产量因子、叶宽因子、每穗小穗数因子、株高因子、叶长因子。其中穗粒数量和重量因子由于方差贡献最大达到14%,此因子对产量的形成起着决定作用。通过改进雷达图综合评价分析得到各样本综合得分,其中前五名的世代依次是:南大2419子五代、南大2419子四代、阿夫子三代、阿夫子四代、南大2419子二代,综合得分最低的是:碧玛4号。由综合得分中三个亲本基因型值和后代估计育种值来看:南大2419亲本在综合得分上贡献给后代的有利增效基因潜力较大;碧玛4号亲本在综合得分上贡献给后代的不利减效基因潜力较大.三个亲本对下代的综合传递能力:碧玛4号最强,南大2419其次,阿夫最弱。
5.通过对2007-08年陕西杨凌矮杆小麦100次模拟小区产量数据的原始线性模型进行有效性分析,结果表明:三个固定效应:重复、处理、重复内区组的多套模拟数据计算的F值中超过50%大于原始数据F值,认为三个固定效应的原始概率值可以接受,即原始线性模型方差分析结果是有效的,可利用模型产生的模拟数据进行分析。通过两个地点数据分析综合说明了三次重复简化广义格子设计分析(SGL)模型能更进一步控制田间试验误差,且增加重复数可以更为精确的估计处理效应,在山东实测数据实验中2次重复(SGL)模型就能达到三次重复随机完全区组分析(RB)模型的试验效率水平。进一步证明了大数量试验中重复次数为3次重复的(SGL)模型能有效提高试验效率。
6.本研究以R语言环境为开发平台以R-Commander为设计模板,遵循软件工程理念,采用模块化设计方法,从研究原有的混合模型分析程序、调整均值估计程序、多重比较方法程序和线性分组比较程序入手,构建了以鼠标、菜单操作的软件分析模块,并初步得到以下结果:(1)构建了混合模型分析模块,利用该模块可用于试验设计中的裂区设计分析,巢式设计分析等,试验者和科研人员可方便的在参数菜单中选择最大似然估计(ML)和限制性最大似然估计(REML)并在固定效应和随机效应参数中对分类因子各水平在不同层次上进行效应估计。(2)增加了调整均值的估计分析模块,方便在以后的协方差分析中对处理因子各水平效应均值进行有效的调整以及在非均衡数据情况下对均衡情况的边缘均值做简单地估计。(3)增加了多重比较分析、线性分组比较模块,并对经典的方法做了扩增(如:加入各种P值调整的多重比较方法和针对任意对比的Scheffe、Tukey线性分组比较法)丰富了试验者和科研人员对多重比较、线性分组比较方法的选择。
Founder parent is a base of improved variety in wheat breeding.A new and good wheat variety cannot be bred or developed without the integrated elite characters of Founder parent.Not only the descendant of Founder Parent were selected wheat varieties with large area as materials in most areas of china.But also It derives a lot of breeding parents who have a widely application.But because of experimental material are made up of founder parents and their progeny. And founder parent and its derivative progeny have a relative affinity.Field trial and the corresponding analytical methods for materials of founder parent and its derivative progeny with basing on the Large quantity and Affinities
similarity which are worth disscussing and researching. This study is consider that the 254 materials of founder parents and their progeny from Yangling dwarf wheat variety trials in 2007-2008.The scheme of design(the design of hierarchical structure) for the number of founder parent and its derivative progeny was determined in prophase.The strain effects and correlation of the material are being solved by using hierarchical mixed model,And significant differences in traits between founder parents and their progeny in different generations was illustrated using methods of Multiple Comparison,Linear Grouped Contrast,Breeding value estimates,Multivariate Statistics analysis and Comprehensive Evaluation.The evolutionary tendency and contribution between founder parents and their progeny has been analyzed more further.Through analyzing the plot yield data which would be partly simulated from YangLing in 2007-2008 of ShanXi Province,and partly measured from TaiAn in 2007-2008 of ShanDong Province.The more relative efficiency and stable models were selected by simulation.Finally,We set R language environment as Development platform,and set R-Commander as Design Template.We constructed a software analytical module.It convenient to analyze and manipulate in the correlation of the material and large number of reginal trials and in the future for experimentalist and research worker.
1. Based on the characteristic affinities similarity in founder parent and its derivative progeny of treatment design. According to the ideas of hierarchical design.“intrablock similarity”and“Relative balance on each derivative generation”should be basic principle during conducted incomplete block design with hierarchical treatment. Using Two-direction main area block Arrange in field, Latin square Arrange in intrablock treatment of main area and Setting center CK,random CK,side CK in main area during Environmental design.
2. The S.E.means on the estimated treatment effects for 14 traits of general linear model and hierarchical mixed model value were 187.58 and 120.89,respectively,the S.E.means on the estimated treatment effects of hierarchical mixed model value was lowered 35.55% than general linear model value. It is observed that hierarchical mixed model on grouping block within repeat could effectively reduce the errors on the estimated treatment effects.Suggestions are useing hierarchical mixed model to estimated treatment effects when it has a Large number of treatments and Data are missing and imbalancing.Through ICC(Interclass Correlation Coefficient),We find that Four traits:plot yield,tillernumber,leaf length,plant height,which has higher ICC values.It shows that individual have higher Similarity within class in 4 traits. according to grouping results by two methods on multiple comparisons.the letter was overlap in Tukey grouping test. It is delicate, Intensive to be Compared on small samplesize.the letter wasn’t overlap in Scott-Knott grouping test.It is succinct,easy to be explained on large samplesize. It is thus suggested that Using the scott-knott method can conduct better the grouping results on large samplesize.
3. Base on 3 founder parents:Funo,Bima4,Nanda2419 and its derivative progeny as research object.then their 12 traits were identified in YangLing of ShanXi province in 2008. In view of the means for 12 traits from founder parent and its first filial generations. Founder parent of Funo can transfer genetic background which have high cumulative level on alleles from 7 traits:plot yield, tillernumber, the ability of stripe rust-resistant, grain number per spike, grain weight per spike, leaf length, leaf width to Its first filial generations. But the cumulative level on alleles from 5 traits: heading stage, spikelet number per spike, 1000 grains weight, ear length, plant height are improved due to the influence of Other parents.
Founder parent of Bima4 can transfer genetic background which have high cumulative level on alleles from 9 traits:plot yield, tillernumber, the ability of stripe rust-resistant, spikelet number per spike, ear length,grain number per spike, grain weight per spike, leaf length, leaf width to Its second filial generations.But the cumulative level on alleles from 3 traits: heading stage, 1000 grains weight, plant height are improved due to the influence of Other parents. Founder parent of Nanda2419 can transfer genetic background which have high cumulative level on alleles from 4 traits:plot yield, tillernumber, spikelet number per spike, leaf width to Its first filial generations.But the cumulative level on alleles from 8 traits: heading stage, the ability of stripe rust-resistant, ear length,grain number per spike, grain weight per spike, leaf length,plant height are improved due to the influence of Other parents. By analysing the 9 traits .It has remained basically steady on 3 founder parents:Funo,Bima4,Nanda2419 and their first filial generations,second filial generations,third filial generations, fourth filial generations, fifth filial generations, It shows that the Founder parent can Steady transfer good genetic background to its derivative progeny.
In view of the breeding value for 12 traits from 3 founder parent:Funo,Bima4,Nanda2419 and its derivative progeny.Founder parent of Funo can transfer favorable genetic effect from 5 traits: the ability of stripe rust-resistant, high grain number per spike, high grain weight per spike, leaf length, leaf width to Its derivative progeny.And stripe rust-resistant, leaf length can transfer more favorable genetic effect to Its derivative progeny.breeding value which have unfavourable genetic effect is small on spikelet number per spike.It shows that Founder parent of Funo can transfer smaller genetic load which have unfavourable genetic effect to Its second filial generations. Founder parent of Bima4 can transfer most favorable and additive genetic effect from 1 traits: tillernumber to Its derivative progeny.But Founder parent of Bima4 can transfer genetic load which have unfavourable genetic effect to Its derivative progeny from 9 traits: low plot yield, Feeling stripe rust, less spikelet number per spike,short ear length,less grain number per spike, light grain weight per spike,short leaf, narrow leaf,high plant height.Founder parent of Nanda2419 can transfer most favorable and beneficiated genetic effect from 4 traits: more spikelet number per spike,high 1000 grains weight,long ear length, heavy grain weight per spike to Its derivative progeny.But Founder parent of Nanda2419 can transfer genetic load which have unfavourable genetic effect to Its derivative progeny from 4 traits: less tillernumber, Feeling stripe rust, short leaf, high plant height.
4. The 17 grouping materials were clustered by using Euclidean distance and group-average method.It suggested that the materials could be divided into 8 groups: GroupⅠis Bima4.GroupⅡis Funo.GroupⅢis inter parents of Bima4.GroupⅣis Nanda2419. GroupⅤare first filial generations, second filial generations, third filial generations, fourth filial generations, fifth filial generations of Funo. GroupⅥare first filial generations, second filial generations, third filial generations, fourth filial generations, fifth filial generations of Nanda2419. GroupⅦis fifth filial generations of Funo. GroupⅧare second filial generations, third filial generations, fourth filial generations of Bima4 respectively.through a Correspondence Analysis, announcing to the characteristic of each groups: The results maximum present Similarity and dissimilarity between Founder parents and its derivative progeny.Through the factor loading was estimated by principal component analysis. The 8 factors will be named:The factor of grain number and weight per spike; The factor of 1000 grains weight; The factor of growth period ; The factor of plot yield; The factor of leaf Width; The factor of spikelet number per spike; The factor of plant height; The factor of leaf length;We find that the factor of grain number per spike and grain weight whose contribution of variance accounted to 14%.This factor play a decisive role in yield-building.Through the comprehensive evaluation and analysis by improved Radar chart.We can get the comprehensive score for each sample. The top 5 generations followed by:the fifth filial generations of Nanda2419,the fourth filial generations of Nanda2419,the third filial generations of Funo,the fourth filial generations of Funo,the second filial generations of Nanda2419. The lowest comprehensive score is Bima4. In view of the Comprehensive scores for genotype value and EBVs from3 founder parent and its derivative progeny.Founder parent of Nanda2419 can transfer more favorable and beneficiated genetic effect to Its derivative progeny. Founder parent of Bima4 can transfer more unfavorable and reductive genetic effect to Its derivative progeny.Comprehensive transmission ability for 3 founder parents and its derivative progeny:The best is Bima4,The next is Nanda2419,The weakest is Funo.
5. Through the effectiveness analysis of the Primitive linear model which based on the 100 times of simulation data for plot yield from Yangling dwarf wheat variety trials in 2007-2008.The results show that Over 50 percent of F value which was calculated by the 100 times of simulation data for Three fixed effects: Repetition, Treatment,Block within Repetition is greater than F value which was calculated by raw data. We think original probability value from three fixed effects is quite acceptable. It is show that the result of ANOVA is effectiveIt is available to produced simulation data by the Primitive linear model, and It can be analyzed.
Through analysis the data from 2 locis, it has been confirmed that the simplified generalized grid design analysis (SGL) is better control the error from field experiment,and It Can be more precise estimateeffects of treatments by increasing the number of Repetition.Test efficiency level in (SGL)model on 2 replications as well as Random complete block designs(RB)model on 3 replications. It is proved that (SGL)model on 3 replications can highly increase test efficiency in large number of experiment.
6. This research based on R language environment platform.And taken R-Commander as the design of the template,established a module of the mixed models.Thismethod could be used in split plot design analysis and Nested design analysis in Design of Experiment.Experimentalist and Research worker can easily choose the maximum likelihood estimation(MLE)and the restricted estimation maximum likelihood (REML) in the parameter menu.And It can estimated the effects for each levels in grouping factors of different hierarchies in Fixed effects and random effects parameters.an increase of estimating adjustment mean modules.It can easily adjust means in each levels of treatment factors for Analysis of covariance in the future,and Marginal Means were simply estimated for assuming the balanced situation in unbalanced data.an increase of Multiple comparison analysis and Linear grouping contrasts modules,and It has increased some methods on the previous(for example:Added some methods of Multiple comparison analysis for P value adjustment and Scheffe、Tukey Linear grouping contrasts for arbitrary contrasts).It enrich the methods of Multiple comparison analysis and Linear grouping contrasts which were selected by Experimentalist and Research worker.
引文
[1]曹廷杰,赵虹,王西成,崔党群,詹克慧.河南省弱春性小麦品种(系)主要农艺性状演变分析[J].河南农业科学,2010,9:13-16.
[2]曹新川,何良荣,熊仁次,郑德明,梅拥军.陆地棉主要性状的对应分析[J].江西棉花.2002,24(5):24-26.
[3]陈国宏.R&D项目中止决策的Fuzzy模式识别[J].科学研究.1998,16(1):68-74.
[4]陈华萍,王照丽,魏育明,郑有良.四川小麦地方品种农艺性状与品质性状的聚类分析[J].麦类作物学报.2006,26(6):29-34.
[5]陈贤.番茄品系的品质性状分析[D].昆明:云南农业大学硕士学位论文,2004.
[6]陈衍泰,陈国宏,李美娟.综合评价方法分类及研究进展[J].管理科学学报,2004,7(2):69-79.
[7]陈应志.国家大豆品种区域试验试验精确度研究[D].北京:中国农业大学硕士学位论文.2004.
[8]董攀,李伟,郑有良.波兰小麦主要农艺性状分析[J].麦类作物学报.2007,27(2):216-222.
[9]董占山.统计分析与MINITAB[J].计算机与农业应用,1996,2:24-35.
[10]杜栋,庞庆华.现代综合评价方法与案例精选[M].北京:清华大学出版社,2005.
[11]范永辉.线性混合效应模型的估计与检验[D].北京:北京工业大学博士学位论文. 2007.
[12]斐喜春,薛河儒.SAS及应用[M].北京:农业出版社,1998.
[13]高惠璇.SAS系统SAS/STAT软件使用手册[M].北京:中国统计出版社, 1997a.
[14]高惠璇.生成SAS数据集的几种方法[J].数理统计与管理,1998, 17(3):49-54.
[15]高惠璇.统计分析系统SAS简介[J].数理统计与管理,1997b,15(2):46-48.
[16]郭鹏军,郭鹏辉,吴晨曦,严慧萍.“雷达图”分析法在医院绩效评价中的应用[J].中国医药导报,2009,22:171-172.
[17]郭亚军.综合评价理论、方法及应用[M].北京:科学出版社,2007.
[18]韩路,贾志宽,韩清芳,王海珍.紫花苜蓿主要性状的对应分析[J].中国草地.2003,25(5):38-42.
[19]何晓群.多元统计分析(第二版)[M].北京:中国人民大学出版社,2008.
[20]何晓群.现代统计分析方法[M].北京:中国人民大学出版社,1998.
[21] Falconer D.S.,Mackay T.F.C.(著),储明星(译),师守堃（??数量遗传学导论[M].北京:中国农业科技出版社,2000.
[22]胡琳,盖钧镒,许为钢,赵新西,张磊,王根松.小麦品质特性的分类及相对重要性分析[J].麦类作物学报.2006,26(5):60-64.
[23]胡希远,高金锋,刘建军.空间效应模型分析田间试验的方法与效果[J].西北农林科技大学学报(自然科学版),2007,35(3):87-92.
[24]胡希远,高金锋.非独立试验数据的一般线性混合模型分析[J].中国农学通报,2007,23(3):121-126.
[25]胡希远.数据非平衡性对试验分析结果的影响[J].西北农林科技大学学报(自然科学版),2004,32(1):103-107.
[26]胡向阳.智能的统计软件-统计专家系统[J].数理统计与管理,1994,13(3):47-50.
[27]金善宝,刘定安.中国小麦品种志[M].农业出版社,1964.
[28]金善宝.中国小麦品种志[M].农业出版社,1986.
[29]金善宝.中国小麦品种志[M].农业出版社,1997.
[30]朗云雯.博苑统计软件三种回归与不完全区组的统计分析[D].昆明:云南农业大学硕士学位论文. 2008.
[31]雷振生,林作楫.河南小麦品种农艺性状演变及今后育种方向[J].中国农业科学.1995.28(A1):28-33.
[32]李红霞,方亮,魏亦勤,裘敏,张双喜,刘旺清,王平.宁夏五十年来春小麦主栽品种农艺性状和品质性状演变[J].种子.2007a.26(3):63-66.
[33]李红霞.宁夏五十年来春小麦主栽品种农艺性状和品质性状演变[J].种子,2007b,26(3):63-66.
[34]李建明.中国小麦骨干亲本(品种)性状演变与遗传差异研究[D].杨凌:西北农林科技大学硕士学位论文,2007:13-21.
[35]李建平,宋喜芳,胡希远.空间效应模型在小麦品种比较试验中的应用[J].麦类作物学报,2008,28 (3) :523–527.
[36]李丽.空间诱变育种甜椒品系试验广义格子设计的效率比较[D].昆明:云南农业大学硕士学位论文,2004.
[37]李世平,张哲夫,安林利,行翠平,韩东翠.临汾号小麦品种主要性状演变及育种目标分析[J].山西农业科学.2000.28(3):6-9.
[38]李小军.小麦骨干亲本碧玛4号的遗传效应分析[D].北京:中国农业科学院作物科学研究所博士学位论文.2009.
[39]李雪胜.利用SAS系统进行数据发掘一从数据到业务优势[J].数理统计与管理,1997, 16(2):46-48.
[40]李燕琛.微机上的SPSS软件包[J].数理统计与管理,1994,13(3):43-47.
[41]林德光.在SAS系统方差分析过程中作统计代换[J].数理统计与管理,1998,16(5):47-52.
[42]凌树洪,李远景,张泽生.双向随机区组设计的研究[J].数理统计与管理.1989.(4):43-55.
[43]刘金刚.SAS软件在质量管理中的应用[J].数理统计与管理,1997,16(6):50-54.
[44]卢纹岱.SPSS for Windows统计分析[M].北京:电子工业出版社,2000.
[45] Montgomery D.C.(著),汪仁官,陈荣昭(译).实验设计与分析[M].北京:中国统计出版社出版,1998.
[46]莫惠栋.农业试验统计(第二版)[M].上海:上海科学技术出版社,1992.
[47]牛玉刚.混合区组试验的设计与分析[J].数理统计与管理.1994.13:19-23.
[48]钱建设,杜肖琳.多元统计分析绘图系统PPCH3.2[J].测井技术,2000,24(4):295-299.
[49]钱曼懋,孙洪伟,宋春华,杨欣明.北部冬小麦品种农艺性状演变研究[J].作物品种资源.1989.(1):3-5.
[50] R.G.D.斯蒂尔,J.H.托里(著).杨纪珂,孙长鸣(译).数理统计的原理和方法—适用于生物科学[M].北京:科学出版社.1979.
[51] Richard A.J.,Dean W.W.(著),陆旋,叶俊(译).实用多元统计分析(第6版)[M].北京:清华大学出版社,2008.
[52]荣廷昭,朱孝达,唐福玉.农业试验与统计分析[M].成都:四川科学技术出版社,1993.
[53]石磊.多水平模型及其统计诊断[M].北京:科学出版社,2008.
[54]四川大学华西公共卫生学院卫生统计学教研室.最新医学统计软件PEMS3.1简介[J].中国医院统计,2005.(3):47.
[55]孙啸,谢建明,周庆.R语言及Bioconductor在基因组分析中的应用[M].北京:科学出版社,2006.
[56]汤旦林,李晓强.读入外部数据并作统计分析[J].数理统计与管理,1997a,16(4):46—48.
[57]汤旦林.SAS数据集[J].数理统计与管理,1997b,16(3):51-49.
[58]汤银才.R语言与统计分析[M].北京:高等教育出版社,2008.
[59]唐启义.实用统计和DPS数据处理系统[M].北京:科技出版社,2002.
[60]田笑明.新疆冬小麦品种更替中农艺性状演变和发展方向的研究.作物学报.1991.17(4):297-303.
[61]童一中.作物育种常用的统计方法[M].上海:上海科学技术出版社,1979.
[62]王德青,万永波,王翔,狄让丽.基于主成分的改进雷达图及其在综合评价中的应用[J].数理统计与管理.2010,29(5):883-889.
[63]王济川,谢海义,姜宝法.多层统计分析模型—方法与应用[M].北京:高等教育出版社,2008,16-18.
[64]王江春,于波,王荣,王丽,耿荣萍,陈永娜,李美玲.山东省小麦品种演变及产量性状的遗传分析[J].山东农业科学.2007.(2):5-7,9.
[65]王军霞,官建成.复合DEA方法在测度企业知识管理绩效中的应用[J].科学研究,2002,(1):84-88.
[66]王群.作物育种大数量品系试验设计与分析[D].昆明:云南农业大学硕士学位论文, 1998.
[67]王松桂.线性混合模型理论的新发展[J].北京工业大学学报.2000,26(3):73-81.
[68]王小明,韩小亮.S-PLUS应用统计教程[M].上海:上海财经大学出版社,2005.
[69]王宗军.综合评价的方法、问题及其研究趋势[J].管理科学学报,1998,1(1):75-79.
[70]韦博成.统计诊断引论[M].南京:东南大学出版社,1991.
[71]吴天侠,盖钧镒,马育华.多品种(系)试验中简化广义格子设计的探讨[J].作物学报,1995b,21(3):300-306.
[72]吴天侠,盖钧镒.植物育种多品种(系)试验中近邻分析的探讨[J].作物学报,1995a,21(1):9-16.
[73]吴天侠.植物育种多品种(品系)试验中简化广义格子设计和近邻分析的探讨[D].南京:南京农业大学博士学位论文. 1990.
[74]夏结来,陈长生,刘贺之,吴冰.中文交互式统计分析软件包NoSA的设计基础功能及应用[J].解放军药学学报,2000.16(4):228-230.
[75]肖枝洪.统计模拟及其R实现[M].武汉:武汉大学出版社,2010.
[76]熊范纶,乔克智,胡海瀛.雄风——专家系统开发工具[M].北京:清华大学出版社,1999.
[77]徐雷,夏结来,陈长生.非典型数据统计分析系统NoSA—基本功能和核心算法[J].数理医药学杂志,2002.l5(3):259-261.
[78]徐绍忠,杨德,廖晓虹,毛孝强.作物育种品系试验四种分析方法比较[J].作物学报,2003,29(1):119-122.
[79]徐绍忠.作物育种品系试验的三种近邻分析方法比较[D].昆明:云南农业大学硕士学位论文.1999.
[80]许世蛟.江苏省小麦品种资源农艺性状演变规律研究[J].湖南农业科学,2009,9:1-3,7.
[81]薛毅,陈立萍.统计建模与R软件[M].北京:清华大学出版社,2007.
[82]杨德.滇型杂交水稻数量遗传分析[M].昆明:云南科技出版社,2002b.
[83]杨德.试验设计与分析[M].北京:中国农业出版社,2002a.
[84]杨铭.Minitab用于中心复合设计与数据处理[J],药学服务与研究,2007.7(3):231-234.
[85]杨文雄.甘肃省春小麦品种演变及有关性状遗传分析[J].西北农业学报.2001.10(2):60-63.
[86]姚国才,姚金保,杨学明,黄胜东,钱存鸣,周朝飞,金德洲.90年代江苏淮南小麦产量性状演变及品质分析[J].南京农专学报. 2000.16(3):13-18.
[87]叶乃兴.日本茶树育种的骨干亲本及其系谱分析[J].中国茶叶.2007,29(5):22-23.
[88]张桂英,张国权,罗勤贵,欧阳韶晖,魏益民.陕西关中小麦品质性状的因子及聚类分析[J].麦类作物学报.2010,30(3):548-554.
[89]张佳进.博苑试验统计软件与R统计平台和数据管理接口设计[D].昆明:云南农业大学硕士学位论文. 2007.
[90]张连平.小麦品种区域试验中品种评价方法的比较[D].北京:中国农业大学硕士学位论文.2006.
[91]张喜征,廖斌,薛啬媛.基于雷达图的企业知识存量测评与预警研究[J].情报杂志,2007,8:2-3,6.
[92]张小燕,宋哲民.陕西关中小麦品种更替中性状演变及其发展方向[J].西北植物学报.1994.14(4):295-302.
[93]赵淑贞,卢金宝,程海卫,支金虎.陆地棉F2代经济性状的对应分析[J].江西棉花.2004,26(6):16-18.
[94]钟声,阮培均.小麦产量及主要产量性状演变的分析[J].种子.1995.(6):24-26.
[95]周吉,徐萍,李国强,柴守玺,张正斌.不同水旱地小麦品种聚类分析及其利用模式研究[J].麦类作物学报.2009,29(5):809-813.
[96]周阳,何中虎,陈新民,王德森,张勇,张改生.30余年来北部冬麦区小麦品种产量改良遗传进展[J].作物学报,2007,33(9):1530-1535.
[97]朱军.线性模型分析原理[M].北京:科学出版社,2000:111-159.
[98]庄巧生.中国小麦品种改良及系谱分析[M].北京:中国农业出版社,2003.
[99] Abdi H.,Edelman B.,Valentin D.,Dowling W.J..Experimental Design and Analysis for Psychology[M].Oxford:Oxford University Press.2009.
[100] Austin R.B.,Bingham J.,Blackwell R.D..Genetic in provements in winter wheat yields since 1900 and associated physiological changes[J].J Agric Sci,1980,94:675-689.
[101] Antunes M.,Sousa L..Bayesian Classification and Non-Bayesian Label Estimation via EM Algorithm to Identify Differentially Expressed Genes: a Comparative Study Biometrical,2008,50(5): 824–836.
[102] Azais J.M.,Monod H.,Bailey R.A..The influence of design on validity and efficiency of neighbour methods[J].Biometrics,1998,54:1374-1387.
[103] Bartlett M.S..The approximate recovery of information from field experiments with large blocks[J].J.Agric.Sci.1938,28:418-427.
[104] Bartlett M.S..Nearest neighbour models in the analysis of field experiments [J].Journal of the Royal Statistical Society.Series B,1978,40:147-174.
[105] Bolker B..Ecological Models and Data in R[M].New Jersey:Princeton University Press,2007.
[106] Benaim A.R. FDR Control by the BH Procedure for Two-Sided Correlated Tests with Implications to Gene Expression Data Analysis.Biometrical,2007,49(1):107-126.
[107] Benjamini Y.,Hochberg Y.. Controlling the false discovery rate: a practical and powerful approach to multiple testing[J].Journal of the Royal Statistical Society Series B,1995,57,289-300.
[108] Benjamini Y.,Yekutieli D..The control of the false discovery rate in multiple testing under dependency[J]. Annals of Statistics 2001,29,1165–1188.
[109] Besag J.,Kempton R..Statistical analysis of field experiments using neighbouring plot[J].Biometrics.1986,42:231-251.
[110] Booker L.,Goldberg D.E.,Holland J.H.Classifier system and genetic algorithms[J].Artificial Intellifence,1989,40(9):1-3.
[111] Bretz F.,Hothorn T.,Westfall P..Multiple Comparisons Using R[M].Boca Raton,Florida USA, 2010.
[112] Brown H.,Prescott R..Applied Mixed Models in Medicine[M]. New Jersey:John Wiley & Sons,2006.
[113] Carrettoni F.,Castano S.,Martella G.,Samaratti P.. RETISS:A real time securitysystem for threat detection using fuzzy logic[A].In:Proceedings of 25th IEEE International Carnahan Conference on Security Technology[C].Taipei:1991.247-269.
[114] Chames A.,Cooper W.W.,Rhodes E..Measuring the efficiency of decision making units[J].European Journal of Operational Research,1978,(2):429-444.
[115] Chen S.J.,Hwang C.L..Fuzzy Multiple Attribute Decision Making[M].Berlin:Springer Press,1992.
[116] Cochran W.G.,Gertrude M.Cox:Experimental Dsigns,2d ed.[M]John Wiley&Sons,Inc.,New York,1957.
[117] Cooper W.W.,Tone K..Measures of inefficiency in data envelopment analysis and stochastic frontier estimation[J].European Journal of Operational Research,1997,(2):72-78.
[118] Dagli C.,Schierholt K..Stock market prediction using different neural network classification architectures[A].In:IEEE/IAFE 1996 Conference on Computational Intelligence for Financial Engineering[C].1996,72-78.
[119] Damon R.A.,Crown R.M.,Singletary C.B.,McCraine S.E..characteristics of purebred and crossbred Beef steers in the Gulf Coast Region[J].journal of animal Science,1959,19,819-844.
[120] David B.,Roger M..The empirical efficiency and validity of Two neighbour models[J].Biometrics, 1991,47:1473-1487.
[121] Demidenko E..Mixed Models:Theory and Applications[M].New Jersey:John Wiley & Sons,2004.
[122] Dimitras A.I.,Slowinski R.,Susmaga R.,Zopounidis C..Business failure prediction using rough sets[J].European Journal of Operational Research,1999,95:24-37.
[123] Duncan D.B..multiple range and multiple F tests[J].Biometrics.1955,11:1-42.
[124] Duncan D.B..t Tests and Intervals for Comparisons Suggested by the Data[J].Biometrics.1975,31(2):339-359.
[125] Dunnett C.W..A multiple comparison procedure for comparing several treatments with a control[J].Journal of the American Statistical Association.1955,50(272):1096-1121.
[126] Efron B.,Tibshirani R.,StoreyJ.D.,Tusher V..Empirical Bayes analysis of a microarray experiment[J].Journal of the American Statistical Association.2001,96(456):1151–1160.
[127] Everitt B.S..An R and S-PLUS Companion to Multivariate Analysis[M].London:Springer-Verlag,2005.
[128] Everitt B.S..Cambridge Dictionary of Statistics 3rdE[M].London:Cambridge University Press,2006.
[129] Faraway J.J..Linear Models with R[M].Boca Raton,Florida USA,Chapman & Hall/CRC,2004.
[130] Federer W.T..Experimental design[M].New York:The Macmillan Company.1955.
[131] Federer W.T..Augumented designs with one-way elimination of heterogeneity[J].Biometrics.1961,17:447-473.
[132] Federer W.T..Augumented(or hoonuiaku)[J].design.HawaiianPlanters’Record.1956,55:191-208.
[133] Federet W.T..Nair,R.C.and Rahavarao,D.Some augumented row-coloumn design[J].Biometrics 1975,31:361-374.
[134] Fisher R.A..The Design of Experiments.Frist Edition[M].Edinburgh and London:Oliver and Boyd.1935.
[135] Fox J..Getting Started with the R Commander[EB/OL].Documentation attached to theRcmdr R-package.2005a[2010-02-27].http://socserv.mcmaster.ca/jfox/Getting-Started-with-the-Rcmdr.pdf
[136] Fox J..The R Commander: A Basic-Statistics Graphical User Interface to R[J]. Journal of Statistical Software,2005b,14(9):1-42.
[137] Gates C.E.,Bilbro J.D..Illustration of a Cluster Analysis Method for Mean Separation[J].Agronomy Journal, 1978,70:462-465.
[138] George A.M.,Dallas E.J..Analysis of Messy Data Volume1:Designed Experiments,SecondEdition[M].Boca Raton:CRC press,2009.
[139] Gerge C.J..Evalution of moving and border row mean covariance analysis for error control in yield trails. J.Ameri.Soc.Hort.Sci.1990,115(21):241-244.
[140] Gomez K.A.,Gomez A.A..Statistical Procedures for Agricultural Resarch[M],The IRRI.,1981.
[141] Goodnight J.H.,Harvey W.R..least squares means in fixed-effects General Linearmodel[R],technical report No.R-103.Raleigh.N.C.:SAS institute.1978.
[142] Goodnight J.H..New features in GLM and VARCOMP[C].Proceeding of the Fourth Annual SAS Users’Group International Conference. Raleigh.N.C.:SAS institute,1979,208-217.
[143] Grabowski M.R.,Wallace W.A..An expert system for maritime pilots:Its design and assessment using gaming[J].Management Science,1993,3(12):1506-1520.
[144] Grob J..A Note on Equality of MINQUE and Simple Estimator in the general Gauss-Markov Model[J].Statistics & Probability Letters.1997,35:335-339.
[145] Gurka M.J..Selecting the Best Linear Mixed Model under REML[J].American statistical Association.2006,60:19-26.
[146] Haiyan Xu, Hsu C. J..Applying the Generalized Partitioning Principle to Control the Generalized Familywise Error Rate.Biometrical,2007,49(1):52-67.
[147] Harvey W.R..least squares analysis of data with unequal subclass numbers[R].Report No.ARS 20-8,A.R.S.,Beltsville,Md:U.S.Department of Agriculture.1960.
[148] Harvey W.R..Instructions for use of LSMLGP (Least-squares and Maximum Likelihood General Purpose Program)[M].Unpublished mimeograph. Columbus: Department of Dairy Science, Ohio State University.1968.
[149] Heiberger R.M.,Holland B.Statistical Analysis and Data Display: An Intermediate Course with Examples in S-Plus/R,and SAS[M].Springer Texts in Statistics. Springer,2004.
[150] Hemmerle W.J.,Hartley H.O..Computing Maximum Likelihood Estimates for the Mixed AOV Model Using the W-Transformation[J].Technometrics,1973,15, 819–831.
[151] Hochberg Y..A sharper Bonferroni procedure for multiple tests of significance[J].Biometrika,1988,75,800-803.
[152] Hocking R.R.,F.M.Speed.A Full Rank Analysis of Some Linear Model Problems[J].Journal of the American Statistical Association,1975,70:706-712.
[153] Holm S..A simple sequentially rejective multiple test procedure[J].Scandinavian Journal of Statistics,1979,6,65-70.
[154] Hommel G..A stagewise rejective multiple test procedure based on a modified Bonferroni test[J].Biometrika,1988,75,383-386.
[155] Hwang C.L.,MdAasud A.S..Multiple Objective Decision-Making Methods and Applications[M].Berlin:Spring-verlag Press,1979.
[156] Shaffer J.P..Controlling the False Discovery Rate with Constraints: The Newman-Keuls Test Revisited.Biometrical,2007,49(1):136–143.
[157] Johnson N.L.,Kotz S.,Balakrishnan N..Continuous univariate distributions(2nd Edition)[M].New York:John Wiley,1995,Vol.2.
[158] Kempton R.A.,Howes C.W..The use of neighbouring plot values in the analysis of variety trials[J].Appl.Stastist.1981,30(1):59-70.
[159] Keuls M..the use of the“studentized range”in connection with an analysis of variance[J].euphytica.1952,1:112-122.
[160] Khan N..Comparison of nearest neighbour analysis with randomized complete block design in large wheat variety evaluation experiment[J],Journal of Agricultural Research,1990,28(2):177-183.
[161] Kramer C.Y..Extension of multiple range test to group means with unequal number of replications[J].Biometrics.1956,12:307-310.
[162] Laird N.M.,Ware J.H..Random-Effects Models for Longitudinal Data[J].Biometrics,1982,38,963-974.
[163] Levine P.,Pomerol J.C..PRIAM,an interactive program for choosing among multiple attribute alternatives[J].European Journal of Operations Research,1986,25(2):272-280.
[164] Lichtenberg F.R..Issues in measuring industrial R&D[J].Research Policy,1990,19(1):157-163.
[165] Lin C.S.,Poushinsky G..Amodified augumented design for an early stage of test line without replication [J].Biometrics .1983a, 39: 533-561.
[166] Lin C.S.,Poushinsky G.,Jui P.Y..Simulation study of three adjustment methods for the modified augumented design and comparison with balanced lattice square design[J].Agric .Sci. Camb.1983b,100: 527-534.
[167] Lindstrom M.J.,Bates D.M..Newton-Raphson and EM Algorithms for Linear Mixed-Effects Models for Repeated-Measures Data[J].Journal of the American Statistical Association,1988,83,1014–1022.
[168] Ma R.H.,Harrington J.B..The standard errors of different designs of field experiment at the university of Saskatchewan[J].Sci.Agr.,1948,28:462-474.
[169] Mak C.J.,Harvey B.L.,Berdnhl J.D..An ecalution of control plots and moving means for error control in Barley nuries[J].Crop Science,1978,18:870-873.
[170] Mendiburu D.F..Package'agricolae'[EB/OL]// Statistical Procedures for Agricultural Research, International Potato Center.University: Agraria La Molina,Peru,2010a[2010-02-27]http://cran.r-project.org/web/packages/agricolae/agricolae.pdf.
[171] Mendiburu D.F..PracticalManualAgricolae[EB/OL]// Statistical Procedures for Agricultural Research, International Potato Center.University: Agraria La Molina,Peru,2010b[2010-02-27]http://tarwi.lamolina.edu.pe/~fmendiburu/index-filer/download/PracticalManualAgricolae.pdf
[172] Michael J.C..The R Book,1st edition[M].London:Wiley Publishing,2007:449-627.
[173] Montgomery D.C..Design and Analysis of Experiments (7th Edition)[M].John Wiley& Sons.2009.
[174] Nakamura T.,Douke H..Development of Sequential Multiple Comparison Procedure for Dose Response Test. Biometrical,2007,49(1):30–39.
[175] Newman D..the distribution of range in samples from a normal population,expressed in terms of an independent estimate of standard deviation[J].Biometrika.1939,31(1-2):20-30.
[176] Handbook of Statistical Methods..Section 7.4.2.6 Assessing the response from any factor combination[EB/OL]Engineering Statistics Handbook.2003[2010-02-27].http://www.itl.nist.gov/div898/handbook/prc/section4/prc426.htm
[177] Papadakis J.S..Methode statistique pour des experiences sur champ[J].Bull.Sci.Inst.Amilioration des Plantes,Thessaloniki,1937,23:13-28.
[178] Paterson L.J.,Patterson H.D..An algorithm for generating alpha-lattice design[J].Ars combination,1983,16A:87-98.
[179] Patterson H.D.,Williams E.R..A new class of resolvable incompete block designs[J].Biometrila,1976,63:83-92.
[180] Patterson H.D.,thompson R..Recovery of inter-block information when block sizes are unequal[J].Biometrika,1971,58(3):545-554.
[181] Patterson H.D.,Hunter E.A..The efficiency of incomplete block design in national list and recommended list cereal variety trials[J].Agric.Sci,1983,101:427-433.
[182] Perry M.W.,D’Antuono M.F..Yield improvement and associated characteristics of some Australian spring wheat cultivars introduced between 1860 and 1982[J].Australian Journal of Agricultural Research,1989,40:457-472.
[183] Pinheiro J.C.,Bates D.M..Mixed-Effects Models in S and S-PLUS[M]Springer,2000.
[184] Pinheiro J.C.,Bates D.M..Unconstrained Parametrizations for Variance-Covariance Matrices[J].Statistics and Computing, 1996,6, 289–296.
[185] Pithuncharurnlap M.,Basford K.E.,Cullis B.R..Nearest neighbour analysis ofunequally replicated trials[J].Australian Journal of Statistics,1992,34(1):1-9.
[186] R Development Core Team..R: A language and environment for statistical computing[EB/OL].R Foundation for Statistical Computing,Vienna, Austria. 2009,(2009-12-14)[2010-01-15], http://cran.r-project.org/doc/manuals/refman.pdf.
[187] Rencher A.C.,Schaalje G.B..Linear models in statistics second edition[M],NewJersey: Wiley Publishing,2008.
[188] Reynolds M.P.,Rajaram S.,Sayre K.D..Physiological and genetic changes of irrigated wheat in the post-green revolution period and approaches for meeting projected global demand[J].Crop Science,1999,39:1611-1621.
[189] Rousseeuw P.J.,Hubert M..Recent developments in PROGRESS[J].In L1-Statistical Procedures and Related Topics ed Y. Dodge,IMS Lecture Notes.1997,31,201–214.
[190] Rousseeuw P.J.,Leroy A.M..Robust Regression and Outlier Detection[M].Wiley,1987.
[191] Rousseeuw P.J.,van Driessen K..A fast algorithm for the minimum covariance determinant estimator[J].Technometrics 1999,41,212–223.
[192] Rousseeuw P.J.,van Zomeren B..Unmasking multivariate outliers and leverage points(with discussion)[J].Journal of the American Statistical Association,1990,85,633-651.
[193] Royo C.,álvaro F.,Martos V.,Ramdani A.,Isidro J.,Villegas D.,García D.,Moral L.F..Genetic changes in durum wheat yield components and associated traits in Italian and Spanish varieties during the 20th century[J].Euphytica,2007,155:259-270.
[194] SASWorkShop.PROC GLM-Handout#2.1..LSMEANS[EB/OL].Statistical Programs-College of Agricultural and Life Sciences University of ldaho.2002[2010-02-27].http://www.uiweb.uidaho.edu/ag/statprog/sas/workshops/glm/handout2.1_glm.pdf
[195] Saty T.L..Fundamentals of Decision Making and Priority Theory with the Analytic Hierarchy Process[M].Princeton:RWS Publications,1994.
[196] Savoy J..Statistical inference in retrieval effectiveness evaluation[J].Information Processing and Management,1997,33(4):495-512.
[197] Scheffe H..A method for judging all contrasts in the analysis of variance[J].Biometrika.1953,40:87-104.
[198] Schen K.S..Avoiding rank reversal in AHP decision-support models[J].European Journal of Operational Research,1994,74:4607-4619.
[199] Scott A.J.,Knott M..A Cluster Analysis Method for Grouping Means in the Analysis of Variance[J].Biometrics,1974,30:507-512.
[200] Searle S.R.,Casella G.,McCulloch C..Variance Components[M].New York:John Wiley& Sons,1992.
[201] Searle S.R.,Speed F.M.,Milliken G.A..Population Marginal Means in the Linear Model: An Alternative to Least Squares Means[J].The American Statistician,1980,34(4):216-221.
[202] Seeger P.,Kjeller M..Efficiency of generalized lattices in Swedish variety trails[J].Swedish Journal of Agricultural Research.1988,18(3):155-159.
[203] Shrout P.E.,Fleiss J.L..Intraclass correlations:Uses in assessing rater reliability[J].Psychological Bulletin.1979,86:420-428.
[204] Slafer G.A.,Andrade F.H..Genetic improvement in bread wheat(Triticum aestivum)yield in Argentina[J].Field Crops Research,1989,21:289-296.
[205] Speed F.M.,Hocking R.R.,Hackney O.P..Methods of Analysis of Linear Models withUnbalanced Data[J].Journal of the American Statistical Association,1978,73,105-112.
[206] Speed F.M.,Hocking R.R..The Use of the R()-Notation with Unbalanced Data[J].The American Statistician,1976,30,30-33.
[207] Steel R.G.D..A class of augmented designs[R].Math.Res.Center Tech.Summary Report.No.56.University of Wisconsin.1958.
[208] Stingh M.,Rai K.N.,Wilkinson G.N..Nearest neighbour analysis of grain yield in pear millet[J].Patancheru,A.P.(India),ICRISAT,1987:18.
[209] Storey J.D.,Tibshirani R..Statistical significance for genome-wide studies[J].PNAS,2003,100(16): 9440–9445.
[210] Stroup W.W.,Baenziger P.S.,Mulitze D.K..Removing spatial variation from wheat yield trials: a comparison of methods[J].Crop Science,1994,86:62-66.
[211] Stroup W.W..Predictable Functions and Prediction Space in the Mixed Model Procedure[J].in Applications of Mixed Models in Agriculture and Related Disciplines,Southern Cooperative Series Bulletin No. 343, Louisiana Agricultural Experiment Station, Baton Rouge,1989,39-48.
[212] Takahashi R..Comparative studies of the fertilizer response of Japanese old new two-rowed cultivars of barley[J].Japan J Breed,1981,31(2):183-198.
[213] Tukey J.W..Comparing Individual Means in the Analysis of Variance[J].Biometrics,1949,5,99-114.
[214] Utz H.F..Comparison on nearest neighbour analysis methods in a series of maizes experiments[J].information science in agriculture–information processing–agriculture science,Ulmer,1990:63-68.
[215] Venables W.N.,Ripley B.D..Modern Applied Statistics with S, 4th Edition[M].Springer-Verlag,2002.
[216] Wallmark J.T..Quality of research measured by citation method and by peer review-a comparison[J].IEEE Trans on Enginerring Management,1986,19(4):218-222.
[217] Wang S.G.,Chow S.C..Advanced Linear Models[M].New York:Marcel Dekker,1994.
[218] Westfall P.H., Troendle J. F..Multiple Testing with Minimal Assumptions.Biometrical,2008,50(5): 745–755.
[219] Wilkinson G.N.,Eckert S.R.,Hancock T.W.,Mayo O..Nearest neighbour(NN)analysis of field experiments[J].Journal of the Royal Statistical Society.Series B,1983,45:151-211.
[220] Williams E.R..A neighbour model for field experiments[J].Biometrika.1986,73(2):279-287.
[221] Williams L.J. A.K.,HervéA..Experimental Design Analysis for Psychology[M].USA:Oxford University Press,2009.
[222] Yates F..A new method of arranging variety trials involving a large number of varieties[J].J.Agric.Sci.1936,26:424-455.
[223] Yates F..The Analysis of Multiple Classifications with Unequal Numbers in the Different Classes[J].Journal of the American statistical association,1934,29:52-66.
[224] Youden W.J..Use of icomplete block replications in estimated tobacco-mosaic virus[J].Contributions from Boce Thompson institute,1937,9:41-48.
[2]曹新川,何良荣,熊仁次,郑德明,梅拥军.陆地棉主要性状的对应分析[J].江西棉花.2002,24(5):24-26.
[3]陈国宏.R&D项目中止决策的Fuzzy模式识别[J].科学研究.1998,16(1):68-74.
[4]陈华萍,王照丽,魏育明,郑有良.四川小麦地方品种农艺性状与品质性状的聚类分析[J].麦类作物学报.2006,26(6):29-34.
[5]陈贤.番茄品系的品质性状分析[D].昆明:云南农业大学硕士学位论文,2004.
[6]陈衍泰,陈国宏,李美娟.综合评价方法分类及研究进展[J].管理科学学报,2004,7(2):69-79.
[7]陈应志.国家大豆品种区域试验试验精确度研究[D].北京:中国农业大学硕士学位论文.2004.
[8]董攀,李伟,郑有良.波兰小麦主要农艺性状分析[J].麦类作物学报.2007,27(2):216-222.
[9]董占山.统计分析与MINITAB[J].计算机与农业应用,1996,2:24-35.
[10]杜栋,庞庆华.现代综合评价方法与案例精选[M].北京:清华大学出版社,2005.
[11]范永辉.线性混合效应模型的估计与检验[D].北京:北京工业大学博士学位论文. 2007.
[12]斐喜春,薛河儒.SAS及应用[M].北京:农业出版社,1998.
[13]高惠璇.SAS系统SAS/STAT软件使用手册[M].北京:中国统计出版社, 1997a.
[14]高惠璇.生成SAS数据集的几种方法[J].数理统计与管理,1998, 17(3):49-54.
[15]高惠璇.统计分析系统SAS简介[J].数理统计与管理,1997b,15(2):46-48.
[16]郭鹏军,郭鹏辉,吴晨曦,严慧萍.“雷达图”分析法在医院绩效评价中的应用[J].中国医药导报,2009,22:171-172.
[17]郭亚军.综合评价理论、方法及应用[M].北京:科学出版社,2007.
[18]韩路,贾志宽,韩清芳,王海珍.紫花苜蓿主要性状的对应分析[J].中国草地.2003,25(5):38-42.
[19]何晓群.多元统计分析(第二版)[M].北京:中国人民大学出版社,2008.
[20]何晓群.现代统计分析方法[M].北京:中国人民大学出版社,1998.
[21] Falconer D.S.,Mackay T.F.C.(著),储明星(译),师守堃（??数量遗传学导论[M].北京:中国农业科技出版社,2000.
[22]胡琳,盖钧镒,许为钢,赵新西,张磊,王根松.小麦品质特性的分类及相对重要性分析[J].麦类作物学报.2006,26(5):60-64.
[23]胡希远,高金锋,刘建军.空间效应模型分析田间试验的方法与效果[J].西北农林科技大学学报(自然科学版),2007,35(3):87-92.
[24]胡希远,高金锋.非独立试验数据的一般线性混合模型分析[J].中国农学通报,2007,23(3):121-126.
[25]胡希远.数据非平衡性对试验分析结果的影响[J].西北农林科技大学学报(自然科学版),2004,32(1):103-107.
[26]胡向阳.智能的统计软件-统计专家系统[J].数理统计与管理,1994,13(3):47-50.
[27]金善宝,刘定安.中国小麦品种志[M].农业出版社,1964.
[28]金善宝.中国小麦品种志[M].农业出版社,1986.
[29]金善宝.中国小麦品种志[M].农业出版社,1997.
[30]朗云雯.博苑统计软件三种回归与不完全区组的统计分析[D].昆明:云南农业大学硕士学位论文. 2008.
[31]雷振生,林作楫.河南小麦品种农艺性状演变及今后育种方向[J].中国农业科学.1995.28(A1):28-33.
[32]李红霞,方亮,魏亦勤,裘敏,张双喜,刘旺清,王平.宁夏五十年来春小麦主栽品种农艺性状和品质性状演变[J].种子.2007a.26(3):63-66.
[33]李红霞.宁夏五十年来春小麦主栽品种农艺性状和品质性状演变[J].种子,2007b,26(3):63-66.
[34]李建明.中国小麦骨干亲本(品种)性状演变与遗传差异研究[D].杨凌:西北农林科技大学硕士学位论文,2007:13-21.
[35]李建平,宋喜芳,胡希远.空间效应模型在小麦品种比较试验中的应用[J].麦类作物学报,2008,28 (3) :523–527.
[36]李丽.空间诱变育种甜椒品系试验广义格子设计的效率比较[D].昆明:云南农业大学硕士学位论文,2004.
[37]李世平,张哲夫,安林利,行翠平,韩东翠.临汾号小麦品种主要性状演变及育种目标分析[J].山西农业科学.2000.28(3):6-9.
[38]李小军.小麦骨干亲本碧玛4号的遗传效应分析[D].北京:中国农业科学院作物科学研究所博士学位论文.2009.
[39]李雪胜.利用SAS系统进行数据发掘一从数据到业务优势[J].数理统计与管理,1997, 16(2):46-48.
[40]李燕琛.微机上的SPSS软件包[J].数理统计与管理,1994,13(3):43-47.
[41]林德光.在SAS系统方差分析过程中作统计代换[J].数理统计与管理,1998,16(5):47-52.
[42]凌树洪,李远景,张泽生.双向随机区组设计的研究[J].数理统计与管理.1989.(4):43-55.
[43]刘金刚.SAS软件在质量管理中的应用[J].数理统计与管理,1997,16(6):50-54.
[44]卢纹岱.SPSS for Windows统计分析[M].北京:电子工业出版社,2000.
[45] Montgomery D.C.(著),汪仁官,陈荣昭(译).实验设计与分析[M].北京:中国统计出版社出版,1998.
[46]莫惠栋.农业试验统计(第二版)[M].上海:上海科学技术出版社,1992.
[47]牛玉刚.混合区组试验的设计与分析[J].数理统计与管理.1994.13:19-23.
[48]钱建设,杜肖琳.多元统计分析绘图系统PPCH3.2[J].测井技术,2000,24(4):295-299.
[49]钱曼懋,孙洪伟,宋春华,杨欣明.北部冬小麦品种农艺性状演变研究[J].作物品种资源.1989.(1):3-5.
[50] R.G.D.斯蒂尔,J.H.托里(著).杨纪珂,孙长鸣(译).数理统计的原理和方法—适用于生物科学[M].北京:科学出版社.1979.
[51] Richard A.J.,Dean W.W.(著),陆旋,叶俊(译).实用多元统计分析(第6版)[M].北京:清华大学出版社,2008.
[52]荣廷昭,朱孝达,唐福玉.农业试验与统计分析[M].成都:四川科学技术出版社,1993.
[53]石磊.多水平模型及其统计诊断[M].北京:科学出版社,2008.
[54]四川大学华西公共卫生学院卫生统计学教研室.最新医学统计软件PEMS3.1简介[J].中国医院统计,2005.(3):47.
[55]孙啸,谢建明,周庆.R语言及Bioconductor在基因组分析中的应用[M].北京:科学出版社,2006.
[56]汤旦林,李晓强.读入外部数据并作统计分析[J].数理统计与管理,1997a,16(4):46—48.
[57]汤旦林.SAS数据集[J].数理统计与管理,1997b,16(3):51-49.
[58]汤银才.R语言与统计分析[M].北京:高等教育出版社,2008.
[59]唐启义.实用统计和DPS数据处理系统[M].北京:科技出版社,2002.
[60]田笑明.新疆冬小麦品种更替中农艺性状演变和发展方向的研究.作物学报.1991.17(4):297-303.
[61]童一中.作物育种常用的统计方法[M].上海:上海科学技术出版社,1979.
[62]王德青,万永波,王翔,狄让丽.基于主成分的改进雷达图及其在综合评价中的应用[J].数理统计与管理.2010,29(5):883-889.
[63]王济川,谢海义,姜宝法.多层统计分析模型—方法与应用[M].北京:高等教育出版社,2008,16-18.
[64]王江春,于波,王荣,王丽,耿荣萍,陈永娜,李美玲.山东省小麦品种演变及产量性状的遗传分析[J].山东农业科学.2007.(2):5-7,9.
[65]王军霞,官建成.复合DEA方法在测度企业知识管理绩效中的应用[J].科学研究,2002,(1):84-88.
[66]王群.作物育种大数量品系试验设计与分析[D].昆明:云南农业大学硕士学位论文, 1998.
[67]王松桂.线性混合模型理论的新发展[J].北京工业大学学报.2000,26(3):73-81.
[68]王小明,韩小亮.S-PLUS应用统计教程[M].上海:上海财经大学出版社,2005.
[69]王宗军.综合评价的方法、问题及其研究趋势[J].管理科学学报,1998,1(1):75-79.
[70]韦博成.统计诊断引论[M].南京:东南大学出版社,1991.
[71]吴天侠,盖钧镒,马育华.多品种(系)试验中简化广义格子设计的探讨[J].作物学报,1995b,21(3):300-306.
[72]吴天侠,盖钧镒.植物育种多品种(系)试验中近邻分析的探讨[J].作物学报,1995a,21(1):9-16.
[73]吴天侠.植物育种多品种(品系)试验中简化广义格子设计和近邻分析的探讨[D].南京:南京农业大学博士学位论文. 1990.
[74]夏结来,陈长生,刘贺之,吴冰.中文交互式统计分析软件包NoSA的设计基础功能及应用[J].解放军药学学报,2000.16(4):228-230.
[75]肖枝洪.统计模拟及其R实现[M].武汉:武汉大学出版社,2010.
[76]熊范纶,乔克智,胡海瀛.雄风——专家系统开发工具[M].北京:清华大学出版社,1999.
[77]徐雷,夏结来,陈长生.非典型数据统计分析系统NoSA—基本功能和核心算法[J].数理医药学杂志,2002.l5(3):259-261.
[78]徐绍忠,杨德,廖晓虹,毛孝强.作物育种品系试验四种分析方法比较[J].作物学报,2003,29(1):119-122.
[79]徐绍忠.作物育种品系试验的三种近邻分析方法比较[D].昆明:云南农业大学硕士学位论文.1999.
[80]许世蛟.江苏省小麦品种资源农艺性状演变规律研究[J].湖南农业科学,2009,9:1-3,7.
[81]薛毅,陈立萍.统计建模与R软件[M].北京:清华大学出版社,2007.
[82]杨德.滇型杂交水稻数量遗传分析[M].昆明:云南科技出版社,2002b.
[83]杨德.试验设计与分析[M].北京:中国农业出版社,2002a.
[84]杨铭.Minitab用于中心复合设计与数据处理[J],药学服务与研究,2007.7(3):231-234.
[85]杨文雄.甘肃省春小麦品种演变及有关性状遗传分析[J].西北农业学报.2001.10(2):60-63.
[86]姚国才,姚金保,杨学明,黄胜东,钱存鸣,周朝飞,金德洲.90年代江苏淮南小麦产量性状演变及品质分析[J].南京农专学报. 2000.16(3):13-18.
[87]叶乃兴.日本茶树育种的骨干亲本及其系谱分析[J].中国茶叶.2007,29(5):22-23.
[88]张桂英,张国权,罗勤贵,欧阳韶晖,魏益民.陕西关中小麦品质性状的因子及聚类分析[J].麦类作物学报.2010,30(3):548-554.
[89]张佳进.博苑试验统计软件与R统计平台和数据管理接口设计[D].昆明:云南农业大学硕士学位论文. 2007.
[90]张连平.小麦品种区域试验中品种评价方法的比较[D].北京:中国农业大学硕士学位论文.2006.
[91]张喜征,廖斌,薛啬媛.基于雷达图的企业知识存量测评与预警研究[J].情报杂志,2007,8:2-3,6.
[92]张小燕,宋哲民.陕西关中小麦品种更替中性状演变及其发展方向[J].西北植物学报.1994.14(4):295-302.
[93]赵淑贞,卢金宝,程海卫,支金虎.陆地棉F2代经济性状的对应分析[J].江西棉花.2004,26(6):16-18.
[94]钟声,阮培均.小麦产量及主要产量性状演变的分析[J].种子.1995.(6):24-26.
[95]周吉,徐萍,李国强,柴守玺,张正斌.不同水旱地小麦品种聚类分析及其利用模式研究[J].麦类作物学报.2009,29(5):809-813.
[96]周阳,何中虎,陈新民,王德森,张勇,张改生.30余年来北部冬麦区小麦品种产量改良遗传进展[J].作物学报,2007,33(9):1530-1535.
[97]朱军.线性模型分析原理[M].北京:科学出版社,2000:111-159.
[98]庄巧生.中国小麦品种改良及系谱分析[M].北京:中国农业出版社,2003.
[99] Abdi H.,Edelman B.,Valentin D.,Dowling W.J..Experimental Design and Analysis for Psychology[M].Oxford:Oxford University Press.2009.
[100] Austin R.B.,Bingham J.,Blackwell R.D..Genetic in provements in winter wheat yields since 1900 and associated physiological changes[J].J Agric Sci,1980,94:675-689.
[101] Antunes M.,Sousa L..Bayesian Classification and Non-Bayesian Label Estimation via EM Algorithm to Identify Differentially Expressed Genes: a Comparative Study Biometrical,2008,50(5): 824–836.
[102] Azais J.M.,Monod H.,Bailey R.A..The influence of design on validity and efficiency of neighbour methods[J].Biometrics,1998,54:1374-1387.
[103] Bartlett M.S..The approximate recovery of information from field experiments with large blocks[J].J.Agric.Sci.1938,28:418-427.
[104] Bartlett M.S..Nearest neighbour models in the analysis of field experiments [J].Journal of the Royal Statistical Society.Series B,1978,40:147-174.
[105] Bolker B..Ecological Models and Data in R[M].New Jersey:Princeton University Press,2007.
[106] Benaim A.R. FDR Control by the BH Procedure for Two-Sided Correlated Tests with Implications to Gene Expression Data Analysis.Biometrical,2007,49(1):107-126.
[107] Benjamini Y.,Hochberg Y.. Controlling the false discovery rate: a practical and powerful approach to multiple testing[J].Journal of the Royal Statistical Society Series B,1995,57,289-300.
[108] Benjamini Y.,Yekutieli D..The control of the false discovery rate in multiple testing under dependency[J]. Annals of Statistics 2001,29,1165–1188.
[109] Besag J.,Kempton R..Statistical analysis of field experiments using neighbouring plot[J].Biometrics.1986,42:231-251.
[110] Booker L.,Goldberg D.E.,Holland J.H.Classifier system and genetic algorithms[J].Artificial Intellifence,1989,40(9):1-3.
[111] Bretz F.,Hothorn T.,Westfall P..Multiple Comparisons Using R[M].Boca Raton,Florida USA, 2010.
[112] Brown H.,Prescott R..Applied Mixed Models in Medicine[M]. New Jersey:John Wiley & Sons,2006.
[113] Carrettoni F.,Castano S.,Martella G.,Samaratti P.. RETISS:A real time securitysystem for threat detection using fuzzy logic[A].In:Proceedings of 25th IEEE International Carnahan Conference on Security Technology[C].Taipei:1991.247-269.
[114] Chames A.,Cooper W.W.,Rhodes E..Measuring the efficiency of decision making units[J].European Journal of Operational Research,1978,(2):429-444.
[115] Chen S.J.,Hwang C.L..Fuzzy Multiple Attribute Decision Making[M].Berlin:Springer Press,1992.
[116] Cochran W.G.,Gertrude M.Cox:Experimental Dsigns,2d ed.[M]John Wiley&Sons,Inc.,New York,1957.
[117] Cooper W.W.,Tone K..Measures of inefficiency in data envelopment analysis and stochastic frontier estimation[J].European Journal of Operational Research,1997,(2):72-78.
[118] Dagli C.,Schierholt K..Stock market prediction using different neural network classification architectures[A].In:IEEE/IAFE 1996 Conference on Computational Intelligence for Financial Engineering[C].1996,72-78.
[119] Damon R.A.,Crown R.M.,Singletary C.B.,McCraine S.E..characteristics of purebred and crossbred Beef steers in the Gulf Coast Region[J].journal of animal Science,1959,19,819-844.
[120] David B.,Roger M..The empirical efficiency and validity of Two neighbour models[J].Biometrics, 1991,47:1473-1487.
[121] Demidenko E..Mixed Models:Theory and Applications[M].New Jersey:John Wiley & Sons,2004.
[122] Dimitras A.I.,Slowinski R.,Susmaga R.,Zopounidis C..Business failure prediction using rough sets[J].European Journal of Operational Research,1999,95:24-37.
[123] Duncan D.B..multiple range and multiple F tests[J].Biometrics.1955,11:1-42.
[124] Duncan D.B..t Tests and Intervals for Comparisons Suggested by the Data[J].Biometrics.1975,31(2):339-359.
[125] Dunnett C.W..A multiple comparison procedure for comparing several treatments with a control[J].Journal of the American Statistical Association.1955,50(272):1096-1121.
[126] Efron B.,Tibshirani R.,StoreyJ.D.,Tusher V..Empirical Bayes analysis of a microarray experiment[J].Journal of the American Statistical Association.2001,96(456):1151–1160.
[127] Everitt B.S..An R and S-PLUS Companion to Multivariate Analysis[M].London:Springer-Verlag,2005.
[128] Everitt B.S..Cambridge Dictionary of Statistics 3rdE[M].London:Cambridge University Press,2006.
[129] Faraway J.J..Linear Models with R[M].Boca Raton,Florida USA,Chapman & Hall/CRC,2004.
[130] Federer W.T..Experimental design[M].New York:The Macmillan Company.1955.
[131] Federer W.T..Augumented designs with one-way elimination of heterogeneity[J].Biometrics.1961,17:447-473.
[132] Federer W.T..Augumented(or hoonuiaku)[J].design.HawaiianPlanters’Record.1956,55:191-208.
[133] Federet W.T..Nair,R.C.and Rahavarao,D.Some augumented row-coloumn design[J].Biometrics 1975,31:361-374.
[134] Fisher R.A..The Design of Experiments.Frist Edition[M].Edinburgh and London:Oliver and Boyd.1935.
[135] Fox J..Getting Started with the R Commander[EB/OL].Documentation attached to theRcmdr R-package.2005a[2010-02-27].http://socserv.mcmaster.ca/jfox/Getting-Started-with-the-Rcmdr.pdf
[136] Fox J..The R Commander: A Basic-Statistics Graphical User Interface to R[J]. Journal of Statistical Software,2005b,14(9):1-42.
[137] Gates C.E.,Bilbro J.D..Illustration of a Cluster Analysis Method for Mean Separation[J].Agronomy Journal, 1978,70:462-465.
[138] George A.M.,Dallas E.J..Analysis of Messy Data Volume1:Designed Experiments,SecondEdition[M].Boca Raton:CRC press,2009.
[139] Gerge C.J..Evalution of moving and border row mean covariance analysis for error control in yield trails. J.Ameri.Soc.Hort.Sci.1990,115(21):241-244.
[140] Gomez K.A.,Gomez A.A..Statistical Procedures for Agricultural Resarch[M],The IRRI.,1981.
[141] Goodnight J.H.,Harvey W.R..least squares means in fixed-effects General Linearmodel[R],technical report No.R-103.Raleigh.N.C.:SAS institute.1978.
[142] Goodnight J.H..New features in GLM and VARCOMP[C].Proceeding of the Fourth Annual SAS Users’Group International Conference. Raleigh.N.C.:SAS institute,1979,208-217.
[143] Grabowski M.R.,Wallace W.A..An expert system for maritime pilots:Its design and assessment using gaming[J].Management Science,1993,3(12):1506-1520.
[144] Grob J..A Note on Equality of MINQUE and Simple Estimator in the general Gauss-Markov Model[J].Statistics & Probability Letters.1997,35:335-339.
[145] Gurka M.J..Selecting the Best Linear Mixed Model under REML[J].American statistical Association.2006,60:19-26.
[146] Haiyan Xu, Hsu C. J..Applying the Generalized Partitioning Principle to Control the Generalized Familywise Error Rate.Biometrical,2007,49(1):52-67.
[147] Harvey W.R..least squares analysis of data with unequal subclass numbers[R].Report No.ARS 20-8,A.R.S.,Beltsville,Md:U.S.Department of Agriculture.1960.
[148] Harvey W.R..Instructions for use of LSMLGP (Least-squares and Maximum Likelihood General Purpose Program)[M].Unpublished mimeograph. Columbus: Department of Dairy Science, Ohio State University.1968.
[149] Heiberger R.M.,Holland B.Statistical Analysis and Data Display: An Intermediate Course with Examples in S-Plus/R,and SAS[M].Springer Texts in Statistics. Springer,2004.
[150] Hemmerle W.J.,Hartley H.O..Computing Maximum Likelihood Estimates for the Mixed AOV Model Using the W-Transformation[J].Technometrics,1973,15, 819–831.
[151] Hochberg Y..A sharper Bonferroni procedure for multiple tests of significance[J].Biometrika,1988,75,800-803.
[152] Hocking R.R.,F.M.Speed.A Full Rank Analysis of Some Linear Model Problems[J].Journal of the American Statistical Association,1975,70:706-712.
[153] Holm S..A simple sequentially rejective multiple test procedure[J].Scandinavian Journal of Statistics,1979,6,65-70.
[154] Hommel G..A stagewise rejective multiple test procedure based on a modified Bonferroni test[J].Biometrika,1988,75,383-386.
[155] Hwang C.L.,MdAasud A.S..Multiple Objective Decision-Making Methods and Applications[M].Berlin:Spring-verlag Press,1979.
[156] Shaffer J.P..Controlling the False Discovery Rate with Constraints: The Newman-Keuls Test Revisited.Biometrical,2007,49(1):136–143.
[157] Johnson N.L.,Kotz S.,Balakrishnan N..Continuous univariate distributions(2nd Edition)[M].New York:John Wiley,1995,Vol.2.
[158] Kempton R.A.,Howes C.W..The use of neighbouring plot values in the analysis of variety trials[J].Appl.Stastist.1981,30(1):59-70.
[159] Keuls M..the use of the“studentized range”in connection with an analysis of variance[J].euphytica.1952,1:112-122.
[160] Khan N..Comparison of nearest neighbour analysis with randomized complete block design in large wheat variety evaluation experiment[J],Journal of Agricultural Research,1990,28(2):177-183.
[161] Kramer C.Y..Extension of multiple range test to group means with unequal number of replications[J].Biometrics.1956,12:307-310.
[162] Laird N.M.,Ware J.H..Random-Effects Models for Longitudinal Data[J].Biometrics,1982,38,963-974.
[163] Levine P.,Pomerol J.C..PRIAM,an interactive program for choosing among multiple attribute alternatives[J].European Journal of Operations Research,1986,25(2):272-280.
[164] Lichtenberg F.R..Issues in measuring industrial R&D[J].Research Policy,1990,19(1):157-163.
[165] Lin C.S.,Poushinsky G..Amodified augumented design for an early stage of test line without replication [J].Biometrics .1983a, 39: 533-561.
[166] Lin C.S.,Poushinsky G.,Jui P.Y..Simulation study of three adjustment methods for the modified augumented design and comparison with balanced lattice square design[J].Agric .Sci. Camb.1983b,100: 527-534.
[167] Lindstrom M.J.,Bates D.M..Newton-Raphson and EM Algorithms for Linear Mixed-Effects Models for Repeated-Measures Data[J].Journal of the American Statistical Association,1988,83,1014–1022.
[168] Ma R.H.,Harrington J.B..The standard errors of different designs of field experiment at the university of Saskatchewan[J].Sci.Agr.,1948,28:462-474.
[169] Mak C.J.,Harvey B.L.,Berdnhl J.D..An ecalution of control plots and moving means for error control in Barley nuries[J].Crop Science,1978,18:870-873.
[170] Mendiburu D.F..Package'agricolae'[EB/OL]// Statistical Procedures for Agricultural Research, International Potato Center.University: Agraria La Molina,Peru,2010a[2010-02-27]http://cran.r-project.org/web/packages/agricolae/agricolae.pdf.
[171] Mendiburu D.F..PracticalManualAgricolae[EB/OL]// Statistical Procedures for Agricultural Research, International Potato Center.University: Agraria La Molina,Peru,2010b[2010-02-27]http://tarwi.lamolina.edu.pe/~fmendiburu/index-filer/download/PracticalManualAgricolae.pdf
[172] Michael J.C..The R Book,1st edition[M].London:Wiley Publishing,2007:449-627.
[173] Montgomery D.C..Design and Analysis of Experiments (7th Edition)[M].John Wiley& Sons.2009.
[174] Nakamura T.,Douke H..Development of Sequential Multiple Comparison Procedure for Dose Response Test. Biometrical,2007,49(1):30–39.
[175] Newman D..the distribution of range in samples from a normal population,expressed in terms of an independent estimate of standard deviation[J].Biometrika.1939,31(1-2):20-30.
[176] Handbook of Statistical Methods..Section 7.4.2.6 Assessing the response from any factor combination[EB/OL]Engineering Statistics Handbook.2003[2010-02-27].http://www.itl.nist.gov/div898/handbook/prc/section4/prc426.htm
[177] Papadakis J.S..Methode statistique pour des experiences sur champ[J].Bull.Sci.Inst.Amilioration des Plantes,Thessaloniki,1937,23:13-28.
[178] Paterson L.J.,Patterson H.D..An algorithm for generating alpha-lattice design[J].Ars combination,1983,16A:87-98.
[179] Patterson H.D.,Williams E.R..A new class of resolvable incompete block designs[J].Biometrila,1976,63:83-92.
[180] Patterson H.D.,thompson R..Recovery of inter-block information when block sizes are unequal[J].Biometrika,1971,58(3):545-554.
[181] Patterson H.D.,Hunter E.A..The efficiency of incomplete block design in national list and recommended list cereal variety trials[J].Agric.Sci,1983,101:427-433.
[182] Perry M.W.,D’Antuono M.F..Yield improvement and associated characteristics of some Australian spring wheat cultivars introduced between 1860 and 1982[J].Australian Journal of Agricultural Research,1989,40:457-472.
[183] Pinheiro J.C.,Bates D.M..Mixed-Effects Models in S and S-PLUS[M]Springer,2000.
[184] Pinheiro J.C.,Bates D.M..Unconstrained Parametrizations for Variance-Covariance Matrices[J].Statistics and Computing, 1996,6, 289–296.
[185] Pithuncharurnlap M.,Basford K.E.,Cullis B.R..Nearest neighbour analysis ofunequally replicated trials[J].Australian Journal of Statistics,1992,34(1):1-9.
[186] R Development Core Team..R: A language and environment for statistical computing[EB/OL].R Foundation for Statistical Computing,Vienna, Austria. 2009,(2009-12-14)[2010-01-15], http://cran.r-project.org/doc/manuals/refman.pdf.
[187] Rencher A.C.,Schaalje G.B..Linear models in statistics second edition[M],NewJersey: Wiley Publishing,2008.
[188] Reynolds M.P.,Rajaram S.,Sayre K.D..Physiological and genetic changes of irrigated wheat in the post-green revolution period and approaches for meeting projected global demand[J].Crop Science,1999,39:1611-1621.
[189] Rousseeuw P.J.,Hubert M..Recent developments in PROGRESS[J].In L1-Statistical Procedures and Related Topics ed Y. Dodge,IMS Lecture Notes.1997,31,201–214.
[190] Rousseeuw P.J.,Leroy A.M..Robust Regression and Outlier Detection[M].Wiley,1987.
[191] Rousseeuw P.J.,van Driessen K..A fast algorithm for the minimum covariance determinant estimator[J].Technometrics 1999,41,212–223.
[192] Rousseeuw P.J.,van Zomeren B..Unmasking multivariate outliers and leverage points(with discussion)[J].Journal of the American Statistical Association,1990,85,633-651.
[193] Royo C.,álvaro F.,Martos V.,Ramdani A.,Isidro J.,Villegas D.,García D.,Moral L.F..Genetic changes in durum wheat yield components and associated traits in Italian and Spanish varieties during the 20th century[J].Euphytica,2007,155:259-270.
[194] SASWorkShop.PROC GLM-Handout#2.1..LSMEANS[EB/OL].Statistical Programs-College of Agricultural and Life Sciences University of ldaho.2002[2010-02-27].http://www.uiweb.uidaho.edu/ag/statprog/sas/workshops/glm/handout2.1_glm.pdf
[195] Saty T.L..Fundamentals of Decision Making and Priority Theory with the Analytic Hierarchy Process[M].Princeton:RWS Publications,1994.
[196] Savoy J..Statistical inference in retrieval effectiveness evaluation[J].Information Processing and Management,1997,33(4):495-512.
[197] Scheffe H..A method for judging all contrasts in the analysis of variance[J].Biometrika.1953,40:87-104.
[198] Schen K.S..Avoiding rank reversal in AHP decision-support models[J].European Journal of Operational Research,1994,74:4607-4619.
[199] Scott A.J.,Knott M..A Cluster Analysis Method for Grouping Means in the Analysis of Variance[J].Biometrics,1974,30:507-512.
[200] Searle S.R.,Casella G.,McCulloch C..Variance Components[M].New York:John Wiley& Sons,1992.
[201] Searle S.R.,Speed F.M.,Milliken G.A..Population Marginal Means in the Linear Model: An Alternative to Least Squares Means[J].The American Statistician,1980,34(4):216-221.
[202] Seeger P.,Kjeller M..Efficiency of generalized lattices in Swedish variety trails[J].Swedish Journal of Agricultural Research.1988,18(3):155-159.
[203] Shrout P.E.,Fleiss J.L..Intraclass correlations:Uses in assessing rater reliability[J].Psychological Bulletin.1979,86:420-428.
[204] Slafer G.A.,Andrade F.H..Genetic improvement in bread wheat(Triticum aestivum)yield in Argentina[J].Field Crops Research,1989,21:289-296.
[205] Speed F.M.,Hocking R.R.,Hackney O.P..Methods of Analysis of Linear Models withUnbalanced Data[J].Journal of the American Statistical Association,1978,73,105-112.
[206] Speed F.M.,Hocking R.R..The Use of the R()-Notation with Unbalanced Data[J].The American Statistician,1976,30,30-33.
[207] Steel R.G.D..A class of augmented designs[R].Math.Res.Center Tech.Summary Report.No.56.University of Wisconsin.1958.
[208] Stingh M.,Rai K.N.,Wilkinson G.N..Nearest neighbour analysis of grain yield in pear millet[J].Patancheru,A.P.(India),ICRISAT,1987:18.
[209] Storey J.D.,Tibshirani R..Statistical significance for genome-wide studies[J].PNAS,2003,100(16): 9440–9445.
[210] Stroup W.W.,Baenziger P.S.,Mulitze D.K..Removing spatial variation from wheat yield trials: a comparison of methods[J].Crop Science,1994,86:62-66.
[211] Stroup W.W..Predictable Functions and Prediction Space in the Mixed Model Procedure[J].in Applications of Mixed Models in Agriculture and Related Disciplines,Southern Cooperative Series Bulletin No. 343, Louisiana Agricultural Experiment Station, Baton Rouge,1989,39-48.
[212] Takahashi R..Comparative studies of the fertilizer response of Japanese old new two-rowed cultivars of barley[J].Japan J Breed,1981,31(2):183-198.
[213] Tukey J.W..Comparing Individual Means in the Analysis of Variance[J].Biometrics,1949,5,99-114.
[214] Utz H.F..Comparison on nearest neighbour analysis methods in a series of maizes experiments[J].information science in agriculture–information processing–agriculture science,Ulmer,1990:63-68.
[215] Venables W.N.,Ripley B.D..Modern Applied Statistics with S, 4th Edition[M].Springer-Verlag,2002.
[216] Wallmark J.T..Quality of research measured by citation method and by peer review-a comparison[J].IEEE Trans on Enginerring Management,1986,19(4):218-222.
[217] Wang S.G.,Chow S.C..Advanced Linear Models[M].New York:Marcel Dekker,1994.
[218] Westfall P.H., Troendle J. F..Multiple Testing with Minimal Assumptions.Biometrical,2008,50(5): 745–755.
[219] Wilkinson G.N.,Eckert S.R.,Hancock T.W.,Mayo O..Nearest neighbour(NN)analysis of field experiments[J].Journal of the Royal Statistical Society.Series B,1983,45:151-211.
[220] Williams E.R..A neighbour model for field experiments[J].Biometrika.1986,73(2):279-287.
[221] Williams L.J. A.K.,HervéA..Experimental Design Analysis for Psychology[M].USA:Oxford University Press,2009.
[222] Yates F..A new method of arranging variety trials involving a large number of varieties[J].J.Agric.Sci.1936,26:424-455.
[223] Yates F..The Analysis of Multiple Classifications with Unequal Numbers in the Different Classes[J].Journal of the American statistical association,1934,29:52-66.
[224] Youden W.J..Use of icomplete block replications in estimated tobacco-mosaic virus[J].Contributions from Boce Thompson institute,1937,9:41-48.
© 2004-2018 中国地质图书馆版权所有 京ICP备05064691号 京公网安备11010802017129号 地址：北京市海淀区学院路29号 邮编：100083 电话：办公室：(+86 10)66554848；文献借阅、咨询服务、科技查新：66554700 |