Likelihood and conditional likelihood inference for generalized additive mixed models for clustered data

Lin and Zhang (J. Roy. Statist. Soc. Ser. B 61 (1999) 381) proposed the generalized additive mixed model (GAMM) as a framework for analysis of correlated data, where normally distributed random effects are used to account for correlation in the data, and proposed to use double penalized quasi-likelihood (DPQL) to estimate the nonparametric functions in the model and marginal likelihood to estimate the smoothing parameters and variance components simultaneously. However, the normal distributional assumption for the random effects may not be realistic in many applications, and it is unclear how violation of this assumption affects ensuing inferences for GAMMs. For a particular class of GAMMs, we propose a conditional estimation procedure built on a conditional likelihood for the response given a sufficient statistic for the random effect, treating the random effect as a nuisance parameter, which thus should be robust to its distribution. In extensive simulation studies, we assess performance of this estimator under a range of conditions and use it as a basis for comparison to DPQL to evaluate the impact of violation of the normality assumption. The procedure is illustrated with application to data from the Multicenter AIDS Cohort Study (MACS).     

Infrared absorption and reflectivity of double perovskite Sr2FeWO6

In this work we present temperature dependent infrared reflectivity and absorption of Sr₂FeWO₆ between 700 and 17 K measured from 40 to 10000 cm⁻¹. The reflectivity spectra show well defined phonon bands peaking at 143, 227, 377 and 625 cm⁻¹ assigned to overlapping vibrational modes split from those active in cubic perovskite. We have also verified that this compound is structurally stable in the whole temperature range and that its optical gap at ∼750 cm⁻¹ (95 meV) undergoes only a minor high temperature decrease ascribed to new thermally accessible levels.     

F-H实验仪与光栅光谱仪连用观测253.7nm谱线

通过F-H实验仪和光栅光谱仪的连用，可实时观测汞原子从6s6s1So态激发到6s6sSP1态，退激时，辐射出能量为4．9eV的光量子，其波长为253．7nm．     

甚高频等离子体增强化学气相沉积法沉积μc-Si∶H薄膜中氧污染的初步研究

对不同的本底真空条件下,采用甚高频等离子体增强化学气相沉积技术沉积的氢化微晶硅（μc_Si∶H）薄膜中的氧污染问题进行了比较研究.对不同氧污染条件下制备的薄膜样品的x射线光电子能谱与傅里叶变换红外吸收光谱测量结果表明：μc_Si∶H薄膜中,氧以Si—O，O—O和O—H三种不同的键合模式存在,不同的键合模式源自不同的物理机理.μc_Si∶H薄膜的Raman光谱、电导率与激活能的测量结果进一步显示：沉积过程中氧污染程度的不同,对μc_Si∶H薄膜的结构特性与电学特性产生显著影响；而不同氧污染对μc_Si∶H薄膜电学特性的影响不同于氢化非晶硅（a_Si:H）薄膜. 关键词： 氢化微晶硅薄膜 甚高频等离子体增强化学气相沉积 氧污染     

1H and 13C NMR assignments of abietane diterpenes from Aegiphila lhotzkyana

An NMR study of one new and several known abietane diterpenes isolated from the roots of Aegiphila lhotzkyana is described. In addition to 1D NMR, several 2D shift‐correlated NMR pulse sequences (¹H–¹H‐COSY, NOESY, HMQC and HMBC) were used to establish all the structures, and unambiguously perform the ¹H and ¹³C chemical shift assignments of the new natural diterpene and three derivatives, the NMR data for which have not been reported previously. Revision of current data assignment for teuvincenone H is also suggested. Copyright © 2003 John Wiley & Sons, Ltd.     

π1-classification of real arrangements with up to eight lines

One of the open questions in the geometry of line arrangements is to what extent does the incidence lattice of an arrangement determine its fundamental group. Line arrangements of up to 6 lines were recently classified by K.M. Fan (Michigan Math. J. 44(2) (1997) 283), and it turns out that the incidence lattice of such arrangements determines the projective fundamental group. We use actions on the set of wiring diagrams, introduced in (Garber et al. (J. Knot Theory Ramf.), to classify real arrangements of up to 8 lines. In particular, we show that the incidence lattice of such arrangements determines both the affine and the projective fundamental groups.     

视频编码新标准H.264中抗误码技术的研究

介绍了视频编码新标准———H .2 6 4中采用的几种抗误码技术 ,基于这些技术 ,新标准确保了压缩视频流在接收端的恢复质量。     

H.323同步控制实现研究

提出并实现了一种H.323多媒体通信系统音视频同步控制算法,算法以正常语音播放为基准,利用H.263码流语法信息,通过调整视频播放帧率实现媒体间同步, 不增加通信冗余度。实际测试结果表明,算法有效地提高了多媒体通信系统音视频媒体同步性能和多媒体的感知服务质量,并已应用于实际多媒体通信系统。     

Asymptotic properties of computationally efficient alternative estimators for a class of multivariate normal models

Parameters of Gaussian multivariate models are often estimated using the maximum likelihood approach. In spite of its merits, this methodology is not practical when the sample size is very large, as, for example, in the case of massive georeferenced data sets. In this paper, we study the asymptotic properties of the estimators that minimize three alternatives to the likelihood function, designed to increase the computational efficiency. This is achieved by applying the information sandwich technique to expansions of the pseudo-likelihood functions as quadratic forms of independent normal random variables. Theoretical calculations are given for a first-order autoregressive time series and then extended to a two-dimensional autoregressive process on a lattice. We compare the efficiency of the three estimators to that of the maximum likelihood estimator as well as among themselves, using numerical calculations of the theoretical results and simulations.