粮食最低收购价政策问题研究

利用聚类分析、逐步回归分析确定小麦与稻谷种植面积模型和指标体系.建立综合指标评价模型和ARMA模型,可知:最低收购政策具有托市效应.结合时间趋势剔除法、蛛网模型,可知:小麦和稻谷价格波动呈现5、3年的周期性,分别满足发散型、收敛型蛛网形态.建立基于补偿法的最低收购价定价模型并用BP神经网络模型进行检验.基于反证法,推知调控最低收购价提高5%小麦种植面积的做法是不可行的.最后提出了调控粮食种植的建议·     

小样本定数截尾场合下威布尔和对数正态分布的拟合检验

本文给出了小样本定数截尾场合下两参数威布尔分布和对数正态分布的拟合检验方法,该方法还适用于一般的位置-刻度参数族分布,论文还通过实际算例说明方法的可行性。     

白骨化尸体死亡时间的推断

本文分析了15具白骨化尸体标本的股骨汞(Hg),铅(Pb),镉(Cd)元素含量数据,在三年的时间内采集了3次,一共收集到45个数据。首先将这组数据看着纵向数据,利用线性随机效应混合模型、Cox随机混合效应模型进行分析,结果显示,如果对每个白骨化尸体标本建立线性模型,可以精确预测出死亡时间,而且不需要采集铅元素含量数据。混合效应模型的预测效果也很好,最大误差不会超过1个月。其次我们对数据不作任何假设,利用机器学习中随机森林方法分析数据,并利用5折交叉验证方法来判断结果的可靠性,训练集和测试集的NMSE分别为0.1205944,0.5604286,因此可以用训练出的模型来预测死亡时间。     

半参数线性混合效应模型的联合变量选择

多数基于线性混合效应模型的变量选择方法分阶段对固定效应和随机效应进行选择,方法繁琐、易产生模型偏差,且大部分非参数和半参数的线性混合效应模型只涉及非参数部分的光滑度或者固定效应的选择,并未涉及非参变量或随机效应的选择。本文用B样条函数逼近非参数函数部分,从而把半参数线性混合效应模型转化为带逼近误差的线性混合效应模型。对随机效应的协方差矩阵采用改进的乔里斯基分解并重新参数化线性混合效应模型,接着对该模型的极大似然函数施加集群ALASSO惩罚和ALASSO惩罚两类惩罚,该法能实现非参数变量、固定效应和随机效应的联合变量选择,基于该法得出的估计量也满足相合性、稀疏性和Oracle性质。文章最后做了个数值模拟,模拟结果表明,本文提出的估计方法在变量选择的准确性、参数估计的精度两个方面均表现较好。     

天然气稳态运行优化的混合整数模型及其算法

天然气稳态运行优化问题的难点在于网络结构复杂、规模大、目标函数及约束高度非线性.针对其混合整数非线性规划模型,基于网络约简和线性化技术,建立了线性近似模型,并提出一种新的求解算法.将新算法用于优化我国西部天然气管网系统,结果表明所提算法是有效的.     

On Moments of Folded and Truncated Multivariate Normal Distributions

Recurrence relations for integrals that involve the density of multivariate normal distributions are developed. These recursions allow fast computation of the moments of folded and truncated multivariate normal distributions. Besides being numerically efficient, the proposed recursions also allow us to obtain explicit expressions of low-order moments of folded and truncated multivariate normal distributions. Supplementary material for this article is available online.     

Sequential Co-Sparse Factor Regression

In multivariate regression models, a sparse singular value decomposition of the regression component matrix is appealing for reducing dimensionality and facilitating interpretation. However, the recovery of such a decomposition remains very challenging, largely due to the simultaneous presence of orthogonality constraints and co-sparsity regularization. By delving into the underlying statistical data-generation mechanism, we reformulate the problem as a supervised co-sparse factor analysis, and develop an efficient computational procedure, named sequential factor extraction via co-sparse unit-rank estimation (SeCURE), that completely bypasses the orthogonality requirements. At each step, the problem reduces to a sparse multivariate regression with a unit-rank constraint. Nicely, each sequentially extracted sparse and unit-rank coefficient matrix automatically leads to co-sparsity in its pair of singular vectors. Each latent factor is thus a sparse linear combination of the predictors and may influence only a subset of responses. The proposed algorithm is guaranteed to converge, and it ensures efficient computation even with incomplete data and/or when enforcing exact orthogonality is desired. Our estimators enjoy the oracle properties asymptotically; a non-asymptotic error bound further reveals some interesting finite-sample behaviors of the estimators. The efficacy of SeCURE is demonstrated by simulation studies and two applications in genetics. Supplementary materials for this article are available online.     

Optimizing gradient conditions in online comprehensive two‐dimensional reversed‐phase liquid chromatography by use of the linear solvent strength model

The linear solvent strength model was used to predict coverage in online comprehensive two‐dimensional reversed‐phase liquid chromatography. The prediction model uses a parallelogram to describe the separation space covered with peaks in a system with limited orthogonality. The corners of the parallelogram are assumed to behave like chromatographic peaks and the position of these pseudo‐compounds was predicted. A mix of 25 polycyclic aromatic compounds were used as a test. The precision of the prediction, span 0–25, was tested by varying input parameters, and was found to be acceptable with root mean square errors of 3. The accuracy of the prediction was assessed by comparing with the experimental coverages. Less than half of experimental coverages were outside prediction ± 1 × root mean square error and none outside prediction ± 2 × root mean square error. Accuracy was lower when retention factors were low, or when gradient conditions affected parameters not included in the model, e.g. second dimension gradient time affects the second dimension equilibration time. The concept shows promise as a tool for gradient optimization in online comprehensive two‐dimensional liquid chromatography, as it mitigates the tedious registration and modeling of all sample constituents, a circumstance that is particularly appealing when dealing with complex samples.     

Evaluation of the quality consistency of powdered poppy capsule extractive by an averagely linear‐quantified fingerprint method in combination with antioxidant activities and two compounds analyses

A novel averagely linear‐quantified fingerprint method was proposed and successfully applied to monitor the quality consistency of alkaloids in powdered poppy capsule extractive. Averagely linear‐quantified fingerprint method provided accurate qualitative and quantitative similarities for chromatographic fingerprints of Chinese herbal medicines. The stability and operability of the averagely linear‐quantified fingerprint method were verified by the parameter r . The average linear qualitative similarity (improved based on conventional qualitative “Similarity”) was used as a qualitative criterion in the averagely linear‐quantified fingerprint method, and the average linear quantitative similarity was introduced as a quantitative one. was able to identify the difference in the content of all the chemical components. In addition, was found to be highly correlated to the contents of two alkaloid compounds (morphine and codeine). A simple flow injection analysis was developed for the determination of antioxidant capacity in Chinese Herbal Medicines, which was based on the scavenging of 2,2‐diphenyl‐1‐picrylhydrazyl radical by antioxidants. The fingerprint–efficacy relationship linking chromatographic fingerprints and antioxidant activities was investigated utilizing orthogonal projection to latent structures method, which provided important pharmacodynamic information for Chinese herbal medicines quality control. In summary, quantitative fingerprinting based on averagely linear‐quantified fingerprint method can be applied for monitoring the quality consistency of Chinese herbal medicines, and the constructed orthogonal projection to latent structures model is particularly suitable for investigating the fingerprint–efficacy relationship.