基于半参数回归模型的最小一乘局部线性算法

根据最小一乘准则,推导出最小一乘局部线性估计的计算方法,并通过对模拟数据的计算和分析,对比最小一乘核算法和最小二乘局部线性算法,验证了最小一乘局部线性算法是一种有效的,稳健的估计方法,并且有降低边界效应的作用．     

Logistic和Lotka-Volterra模型参数的灰色估计方法研究

根据灰色系统建模原理,对Logistic模型中的参数进行了估计,将此方法应用于Gause(1934)草履虫实验数据的拟合,并与已有拟合结果对比,说明此方法是有效的.然后将其推广至两种群的Lotka—Volterra模型参数的确定,来研究山西省人口—经济系统的发展演化状况,得出了满意的结果.     

Empirical likelihood for AR-ARCH models based on LAD estimation

By employing the empirical likelihood method,confidence regions for the stationary AR(p)-ARCH(q) models are constructed.A self-weighted LAD estimator is proposed under weak moment conditions.An empirical log-likelihood ratio statistic is derived and its asymptotic distribution is obtained.Simulation studies show that the performance of empirical likelihood method is better than that of normal approximation of the LAD estimator in terms of the coverage accuracy,especially for relative small size of observation.     

农残水平测试中数据多态性分析及指定值估计研究

对全国农残水平测试中毒死蜱、氯氰菊酯数据进行统计分析,在数据统计分布特征研究基础上,使用内核密度估计进行数据多态性分析,使用bootstrap模拟取样法对数据样本值重复取样,以获得稳健的水平测试样品待测物含量代表值估计、标准误差及置信区间描述,证明以bootstrap模拟取样法获取的均值与标准偏差作为有限单次样本代表值是合理、有效的,解决了四分位稳健统计方法对非正态多态分布代表值估计不稳定问题及取样理论中取样样本数限制的瓶颈,为能力验证计划指定值的获取提供了一种新方法。     

Analysis of large experimental datasets in electrochemical impedance spectroscopy

An approach for the analysis of large experimental datasets in electrochemical impedance spectroscopy (EIS) has been developed. The approach uses the idea of successive Bayesian estimation and splits the multidimensional EIS datasets into parts with reduced dimensionality. Afterwards, estimation of the parameters of the EIS-models is performed successively, from one part to another, using complex nonlinear least squares (CNLS) method. The results obtained on the previous step are used as a priori values (in the Bayesian form) for the analysis of the next part. To provide high stability of the sequential CNLS minimisation procedure, a new hybrid algorithm has been developed. This algorithm fits the datasets of reduced dimensionality to the selected EIS models, provides high stability of the fitting and allows semi-automatic data analysis on a reasonable timescale. The hybrid algorithm consists of two stages in which different zero-order optimisation strategies are used, reducing both the computational time and the probability to overlook the global optimum. The performance of the developed approach has been evaluated using (i) simulated large EIS dataset which represents a possible output of a scanning electrochemical impedance microscopy experiments, and (ii) experimental dataset, where EIS spectra were acquired as a function of the electrode potential and time. The developed data analysis strategy showed promise and can be further extended to other electroanalytical EIS applications which require multidimensional data analysis.     

Computing the conditioning of the components of a linear least‐squares solution

In this paper, we address the accuracy of the results for the overdetermined full rank linear least‐squares problem. We recall theoretical results obtained in (SIAM J. Matrix Anal. Appl. 2007; 29 (2):413–433) on conditioning of the least‐squares solution and the components of the solution when the matrix perturbations are measured in Frobenius or spectral norms. Then we define computable estimates for these condition numbers and we interpret them in terms of statistical quantities when the regression matrix and the right‐hand side are perturbed. In particular, we show that in the classical linear statistical model, the ratio of the variance of one component of the solution by the variance of the right‐hand side is exactly the condition number of this solution component when only perturbations on the right‐hand side are considered. We explain how to compute the variance–covariance matrix and the least‐squares conditioning using the libraries LAPACK (LAPACK Users' Guide (3rd edn). SIAM: Philadelphia, 1999) and ScaLAPACK (ScaLAPACK Users' Guide. SIAM: Philadelphia, 1997) and we give the corresponding computational cost. Finally we present a small historical numerical example that was used by Laplace (Théorie Analytique des Probabilités. Mme Ve Courcier, 1820; 497–530) for computing the mass of Jupiter and a physical application if the area of space geodesy. Copyright © 2008 John Wiley & Sons, Ltd.     

Spherical harmonics MUSIC versus conventional MUSIC

With the three-dimensional symmetry and wide potential application, spherical array signal processing has been a hot research area for years. This paper devotes to the direction-of-arrival (DOA) estimation of the spherical arrays. Based on the orthogonality of the sensors’ location, MUSIC algorithm in spherical space is proposed, named as SH-MUSIC. Similar to beamspace MUSIC, spherical harmonics transformation is operated before MUSIC algorithm and a better performance is gotten because SH-MUSIC utilizes the array configuration’s orthogonality. On account of the transformation matrix’s orthogonality, spherical harmonics transformation is suggested to be operated firstly in other improved MUSIC algorithms without rejection, and it is demonstrated in beamspace MUSIC. In addition, owing to the tiny error between the steering vectors and the spherical harmonics with high order, sphere array data models including open sphere and rigid sphere are constructed. Simulation proves SH-MUSIC to be effective. Moreover, experimental data from a rigid sphere microphone array is dealt with by SH-MUSIC and the DOAs are estimated accurately.     

Morphology Development in Multi‐Component Polymer Blends: I Composition Effect on Phase Morphology in PP/PET Polymer Blends

The relationship of the phase morphology of polypropylene/polyethylene‐terephthalate (PP/PET) blends and their corresponding compatibilized blends with composition was investigated using digital image analysis. A diameter, d _g , was defined and calculated to discuss the phase morphology of this polymer blend system. A figure‐estimation method was introduced to determine the width of the distribution of d _g . Based on the method, it is proven that the distribution of d _g obeys a log‐normal distribution and consequently, the distribution width, σ was calculated. Further, a fractal dimension, D _f , was introduced to describe the distribution of main sizes of the particles of the dispersed phase. The results showed that, while d _g increased with the concentration of the dispersed phase, σ and D _f show different dependence relations on composition;σ increases monotonously but D _f shows a maximum at a PET content of 30%, indicating that, even though the whole size distribution is much broader, the distribution of the main body of size becomes more uniform when the content of PET is less than 30%.     

Precise Radiometry: Some Recent Aspects of Fruitful Interaction with Atomic Physics

Modern radiometric analytics demands a complex consideration of nuclear and electron shell processes, if more pretentious aims are envisaged. As an example the small variation of decay rates of radionuclides presents possibilities for information on chemical situations of decaying atoms. In principle this phenomenon is well known since many years, but now the situation is such that, e.g. in ^99mTc internal conversion, a full agreement of the difficult experiments and the respective theory was established. The secondary emission of X-rays as a consequence of high excitation of electron shells in combination with nuclear transitions supplies another example for a methodical progress of radiometry. Investigations on ⁵¹Cr as an electron capture nuclide have shown that chemically induced variations of the K_α to K_β X-ray intensity ratio is at least qualitatively understood.     

Multiplicative processes and power laws in human reaction times derived from hyperbolic functions

In sensory psychophysics reaction time is a measure of the stochastic latency elapsed from stimulus presentation until a sensory response occurs as soon as possible. A random multiplicative model of reaction time variability is investigated for generating the reaction time probability density functions. The model describes a generic class of hyperbolic functions by Piéron?s law. The results demonstrate that reaction time distributions are the combination of log-normal with power law density functions. A transition from log-normal to power law behavior is found and depends on the transfer of information in neurons. The conditions to obtain Zipf?s law are analyzed.