Interval estimation of the critical value in a general linear model

Summary This paper concerns interval estimation of the critical value θ which satisfies under the general linear model,Y _i =μ(x _i )+ε _i (i=1,2,···), where for and the functional forms off _j ^′ s are known. From an asymptotic expansion it is shown that, under reasonable conditions, the limiting distribution of is normal. Thus in the large-sample case a confidence interval for θ can be obtained. Such a result is useful when one is interested in carrying out a retrospective analysis rather than designing the experiment (as in the Kiefer-Wolfowitz procedure). In Section 3 a sequential procedure is considered for confidence intervals with fixed width 2d. It is shown that, for a given stopping variableN, is also asymptotically normal asd→0. Thus the coverage probability converges to 1−α (preassigned) asd→0. An example of application in estimating the phase parameter in circadian rhythms is given for the purpose of illustration. Research partially supported by the NSF Grant DMS-8502346.     

Comments on the primary and scatter dose-spread kernels used for convolution methods

The basic primary and scatter dose-spread kernels used for convolution methods are usually produced by Monte Carlo simulations with the interaction point forced to the center of a large water phantom. However, it is still not clear whether such Monte Carlo based kernels allow accurate dose calculations with a wide range of field sizes and depths, especially in thorax phantoms. Using the differential primary and scatter concept, this paper proposes another type of basic kernel, with which perfectly accurate primary and scatter absorbed dose calculations can be performed under conditions that the beam is parallel, the incident beam intensity is uniform within and zero outside the field, and the primary beam attenuation coefficient along raylines is not a function of depth and off-axis distance.     

Polyparametric modification equation for estimating thermodynamic properties of energetic nitro compounds

Calculation methods, based on hybrid density-functional theory with the basis sets of B3LYP/ 6-311+G (2d, p)//B3LYP/6-31G(d, p)and B3LYP/6-31+G(d)//B3LYP/6-31G(d, p), were applied to determine the thermodynamic characteristics of various energetic nitro compounds. A parametric modification equation and the least-squares approach were used to identify 21 of the energetic research compounds. The atomization energies of these 21 compounds have an average relative error of 0.21–0.25% of the experimental values. The enthalpy (H _f) and the Gibbs energy (G _f) of formation have mean absolute errors of 10.8–11.4 kJ/ mol (2.6–2.7 kcal/mol) and 10.0–10.3 kJ/mol (2.4 kcal/ mol), respectively. The enthalpy and the Gibbs energy of formation obtained exceed those in the literature obtained by semiempirical calculations. The calibrated least-squares parameters and parametric equations were used to predict H _f and G _f for the five newly developed energetic nitro compounds for further applications.Acknowledgements. The authors would like to thank the National Science Council of the Republic of China for financial support of this work under grant no. NSC-91-2113-M-014-003. The National Center for High-Performance Computing providing the computation facility is also acknowledged.     

病态分析体系有偏估计的研究

运用广义岭估计和Ｌｉｕｋｅｊｉａｎ提出的有偏估计，对病态分析体系进行了数值模拟和实际光度测定，结果表明，广义岭估计显优于最小二乘估计，Ｌｉｕｋｅｊｉａｎ法有功效，可和为解析病态分析体系的化学计量学方法。     

Parameter identification of thermophilic anaerobic degradation of valerate

The considered mathematical model of the decomposition of valerate presents three unknown kinetic parameters, two unknown stoichiometric coefficients, and three unknown initial concentrations for biomass. Applying a structural identifiability study, we concluded that it is necessary to perform simultaneous batch experiments with differenitial conditions for estimating these parameters. Four simultaneous batch experiments were conducted at 55°C, characterized by four different initial acetate concentrations. Product inhibition of valerate degradation by acetate was considered. Practical identification was done optimizing the sum of the multiple determination coefficients for all measured state viariables and for all experiments simultaneously. The estimated values of kinetic parameters and stoichiometric coefficients were characterized by the parameter correlation matrix, the confidence interval, and the student's t-test at 9% significance level with positive results except for the saturation constant, for which more eperiments for improving its identifiability should be conducted. In this article, we discussekinetic parameter estimation methods.     

曲率流的参数化有限元逼近

许多物理现象可以在数学上描述为受曲率驱动的自由界面运动,例如薄膜和泡沫的演变、晶体生长,等等.这些薄膜和界面的运动常依赖于其表面曲率,从而可以用相应的曲率流来描述,其相关自由界面问题的数值计算和误差分析一直是计算数学领域中的难点.参数化有限元法是曲率流的一类有效计算方法,已经能够成功模拟一些曲面在几类基本的曲率流下的演化过程.本文重点讨论曲率流的参数化有限元逼近,它的产生、发展和当前的一些挑战.     

Integral Operators Between Fock Spaces∗

In this paper, the authors study the integral operator Sφf(z) = Z C φ(z, w)f(w)dλα(w) induced by a kernel function φ(z, ·) ∈ F ∞α between Fock spaces. For 1 ≤ p ≤ ∞, they prove that Sφ : F 1 α → F p α is bounded if and only if sup a∈C kSφkakp,α < ∞, (?) where ka is the normalized reproducing kernel of F 2 α; and, Sφ : F 1 α → F p α is compact if and only if lim |a|→∞ kSφkakp,α = 0. When 1 < q ≤ ∞, it is also proved that the condition (?) is not sufficient for boundedness of Sφ : F q α → F p α . In the particular case φ(z, w) = eαzw?(z ? w) with ? ∈ F 2 α, for 1 ≤ q < p < ∞, they show that Sφ : F p α → F q α is bounded if and only if ? = 0; for 1 < p ≤ q < ∞, they give sufficient conditions for the boundedness or compactness of the operator Sφ : F p α → F q α.     

高维稳健主成分聚类方法及其应用研究

随着信息技术的高速发展,每条数据所包含的信息越来越丰富,使得数据不可避免地含有异常值,且随着维数的增加,异常值出现的可能性更大。传统的主成分聚类分析对异常值特別敏感,基于MCD估计的主成分聚类方法虽然对异常值具有防御作用,但是在高维数据下MCD估计的偏差过大,其稳健性显著降低,而且当维数大于观测值个数时MCD估计失效。为此本文提出了基于MRCD估计的稳健主成分聚类方法,数值模拟和实证分析表明,基于MRCD估计的主成分聚类分析的效果优于传统的主成分聚类分析和基于MCD估计的主成分聚类分析,尤其是在维数大于样本观测值的情况下,MRCD估计更为有效。     

基于新型核与灰度序列的时滞GM(1,N)模型及其应用

为了解决GM(1,N)模型在新型核与灰度的基础上,对驱动项的延迟作用机理不明确的问题,将时滞参数引入到GM(1,N)模型的驱动项中,构建了基于新型核与灰度的时滞GM(1,N)模型,分析了时滞参数的辨识方法,讨论了新模型的建模机理。为了更好地对该模型的有效性进行验证,将优化的时滞GM(1,N)模型对南京市的雾霾进行预测分析,选择GM(1,N)模型、一元回归模型与文中的优化模型进行对比。结果显示,优化模型对PM₁₀浓度的拟合精度更高,且误差均控制在5%之内,从而验证了提出的优化模型适用于具有时滞特征数据的模拟和预测。     

Weighted local polynomial estimations of a non-parametric function with censoring indicators missing at random and their applications

In this paper, we consider the weighted local polynomial calibration estimation and imputation estimation of a non-parametric function when the data are right censored and the censoring indicators are missing at random, and establish the asymptotic normality of these estimators. As their applications, we derive the weighted local linear calibration estimators and imputation estimations of the conditional distribution function, the conditional density function and the conditional quantile function, and investigate the asymptotic normality of these estimators. Finally, the simulation studies are conducted to illustrate the finite sample performance of the estimators.