EMPIRICAL BAYES ESTIMATION FOR ESTIMABLE FUNCTION OF REGRESSION COEFFICIENT IN A MULTIPLE LINEAR REGRESSION MODEL

In this paper we consider the empirical Bayes (EB) estimation problem for estimable function of regression coefficient in a multiple linear regression model Y=Xβ+e. where e with given β has a multivariate standard normal distribution. We get the EB estimators by using kernel estimation of multivariate density function and its first order partial derivatives. It is shown that the convergence rates of the EB estimators are under the condition where an integer k > 1 . is an arbitrary small number and m is the dimension of the vector Y.

NON-LINEAR DYNAMIC MODEL RETRIEVAL OF SUBTROPICAL HIGH BASED ON EMPIRICAL ORTHOGONAL FUNCTION AND GENETIC ALGORITHM

Aiming at the difficulty of accurately constructing the dynamic model of subtropical high, based on the potential height field time series over 500 hPa layer of T106 numerical forecast products, by using EOF（empirical orthogonal function） temporal-spatial separation technique, the disassembled EOF time coefficients series were regarded as dynamical model variables, and dynamic system retrieval idea as well as genetic algorithm were introduced to make dynamical model parameters optimization search, then, a reasonable non-linear dynamic model of EOF time-coefficients was established. By dynamic model integral and EOF temporal-spatial components assembly, a mid-/long-term forecast of subtropical high was carried out. The experimental results show that the forecast results of dynamic model are superior to that of general numerical model forecast results. A new modeling idea and forecast technique is presented for diagnosing and forecasting such complicated weathers as subtropical high.

A high wind geophysical model fuction for Quik SCAT wind retrievals and application to Typhoon IOKE

The geophysical model function （GMF） describes the relationship between a backscattering and a sea surface wind, and enables a wind vector retrieval from backscattering measurements. It is clear that the GMF plays an important role in an ocean wind vector retrieval. The performance of the existing Ku-band model function QSCAT-1 is considered to be effective at low and moderate wind speed ranges. However, in the conditions of higher wind speeds, the existing algorithms diverge alarmingly, owing to the lack of in situ data required for developing the GMF for the high wind conditions, the QSCAT-1 appears to overestimate the a0, which results in underestimating the wind speeds. Several match-up QuikSCAT and special sensor microwave/imager （SSM/I） wind speed measurements of the typhoons occurring in the west Pacific Ocean are analyzed. The results show that the SSM/I wind exhibits better agreement with the ＂best track＂ analysis wind speed than the QuikSCAT wind retrieved using QSCAT-1. On the basis of this evaluation, a correction of the QSCAT-1 model function for wind speed above 16 m/s is proposed, which uses the collocated SSM/I and QuikSCAT measurements as a training set, and a neural network approach as a multiple nonlinear regression technologytechnology.In order to validate the revised GMF for high winds, the modified GMF was applied to the QuikSCAT observations of Hurricane IOKE. The wind estimated by the QuikSCAT for Typhoon IOKE in 2006 was improved with the maximum wind speed reaching 55 m/s. An error analysis was performed using the wind fields from the Holland model as the surface truth. The results show an improved agreement with the Holland model wind when compared with the wind estimated using the QSCAT-1. However, large bias still existed, indicating that the effects of rain must be considered for further improvement.

A Gravity Forward Modeling Method based on Multiquadric Radial Basis Function

It is one of the most important part to build an accurate gravity model in geophysical exploration.Traditional gravity modelling is usually based on grid method,such as difference method and finite element method widely used.Due to self-adaptability lack of division meshes and the difficulty of high-dimensional calculation.

Empirical Likelihood Inference for Generalized Partially Linear Models with Longitudinal Data

In this article, we propose a generalized empirical likelihood inference for the parametric component in semiparametric generalized partially linear models with longitudinal data. Based on the extended score vector, a generalized empirical likelihood ratios function is defined, which integrates the within-cluster?correlation meanwhile avoids direct estimating the nuisance parameters in the correlation matrix. We show that the proposed statistics are asymptotically?Chi-squared under some suitable conditions, and hence it can be used to construct the confidence region of parameters. In addition, the maximum empirical likelihood estimates of parameters and the corresponding asymptotic normality are obtained. Simulation studies demonstrate the performance of the proposed method.

Response of the North Pacific Oscillation to global warming in the models of the Intergovernmental Panel on Climate Change Fourth Assessment Report

Based on 22 of the climate models from phase 3 of the Coupled Model Intercomparison Project, we investigate the ability of the models to reproduce the spatiotemporal features of the wintertime North Pacific Oscillation(NPO), which is the second most important factor determining the wintertime sea level pressure field in simulations of the pre-industrial control climate, and evaluate the NPO response to the future most reasonable global warming scenario(the A1B scenario). We reveal that while most models simulate the geographic distribution and amplitude of the NPO pattern satisfactorily, only 13 models capture both features well. However, the temporal variability of the simulated NPO could not be significantly correlated with the observations. Further analysis indicates the weakened NPO intensity for a scenario of strong global warming is attributable to the reduced lower-tropospheric baroclinicity at mid-latitudes, which is anticipated to disrupt large-scale and low-frequency atmospheric variability, resulting in the diminished transfer of energy to the NPO, together with its northward shift.

基于遗传算法的磨削力模型系数优化及验证

在磨削力模型求解问题中,目前大多使用分段计算法或列方程组直接计算各个待求系数,不仅计算量大且其精度也无法保证。另外,传统的回归模型容易陷入局部最优,难以描述非线性关系。为此,将遗传算法引入到非线性优化函数参数优化中,基于外圆横向磨削力模型、平面磨削力模型、外圆纵向磨削力模型等现有的模型数据,开展磨削力理论模型的系数优化方法研究。相关性分析结果表明:通过计算得到的3种模型磨削力的预测精度提高了14.69%~42.54%,且3种模型所预测的法向磨削力的平均误差分别为5.9%、9.13%、3.23%,切向力平均误差分别为6.78%、8.36%、3.69%。经对比知,优化后的模型拟合度较好,模型预测精度显著提高。遗传算法优化后的非线性优化函数GA-LSQ算法更适合磨削力模型的求解,可对磨削力的预测及实际加工生产中的参数优化提供参考。

一种最新的经验投影函数GMF分析

介绍一种可以在全球范围使用的投影函数GMF,由GMF计算得到的对流层延迟分别与分段线性估计的延迟及NMF模型的数值分析结果比较表明,GMF是一种精度更高且易于软件实现的经验投影函数,具有很好的实用价值。

EEMD-小波在高边坡变形信息提取中的应用研究

针对高边坡变形呈现非平稳性及数据“噪声”多源的问题,提出了一种定向滤波的变形信息提取方法。首先,利用集合经验模态分解方法分解变形时序数据,结合定量分析法判别模态分量信号频段;然后,对高频模态分量中的“噪声”利用小波函数进行“靶向”消噪处理,并对趋势项进行傅里叶级数拟合;最后,重构高边坡变形分析模型,实现真实变形量的提取。结果表明,对比分析各项检验指标,通过“靶向”消噪,各高频模态分量消噪效果明显,重构后的集合经验模态分解(Ensemble Empirical Mode Decomposition,EEMD)-小波高边坡变形分析模型较原始形变和其他模型在精度指标方面提升显著,该方法可用于高边坡的变形预测分析和真实变形量提取。

基于GMF方法的捕风一号风速反演分析

全球尺度高时空分辨率海面风场探测是当前全球气象研究及预报预测领域的关注热点之一,传统海面风场探测技术存在测量区域有限,且受天气环境限制明显等问题。基于全球导航卫星系统-反射(GNSS-R)测量技术风速反演原理,以捕风一号1级数据产品为输入,欧洲中期天气预报中心再分析风速数据为参考风速,采用地球物理模型函数(GMF)方法构建风速反演模型。分析了1级产品中不同参量,如卫星、观测天线、镜面点入射角、信噪比等,对观测特征量与风速的影响,确定了GMF经验模型分类和筛选方案并完成模型建立。对所生成的风速反演数据及其统计特性与飓风卫星导航系统(CYGNSS)相应数据进行了比对分析,表明反演风速以及反演偏差关于入射角、参考风速的分布特性均与CYGNSS数据趋势一致。初步展示了捕风一号卫星风速探测能力,可为后续探测性能提升和星座发展提供参考。

Testing for Error Correlation in Semi-Functional Linear Models

Existing methods for analyzing semi-functional linear models usually assumed that random errors are not serially correlated or serially correlated with the known order.However,in some applications,these assumptions on random errors may be unreasonable or questionable.To this end,this paper aims at testing error correlation in a semi-functional linear model(SFLM).Based on the empirical likelihood approach,the authors construct an empirical likelihood ratio statistic to test the serial correlation of random errors and identify the order of autocorrelation if the serial correlation holds.The proposed test statistic does not need to estimate the variance as it is data adaptive and possesses the nonparametric version of Wilks'theorem.Simulation studies are conducted to investigate the performance of the proposed test procedure.Two real examples are illustrated by the proposed test method.

Modeling the global ionospheric total electron content with empirical orthogonal function analysis

In the present work we model the global ionospheric total electron content (TEC) with the analysis of empirical orthogonal functions (EOF). The obtained statistical eigen modes, which makeup the modeled TEC, consist of two factors: the eigen vectors mapping TEC patterns at latitude and longitude (or local time LT), and the corresponding coefficients displaying the TEC variations in different time scales, i.e., the solar cycle, the yearly (annual and semiannual) and the diurnal universal time variations. It is found that the EOF analysis can separate the TEC variations into chief processes and the first two modes illustrate the most of the ionospheric climate properties. The first mode contains both the semiannual component which shows the semiannual ionospheric anomaly and the annual component which shows the annual or non-seasonal ionospheric anomaly. The second mode contains mainly the annual component and shows the normal seasonal ionospheric variation at most latitudes and local time sectors. The annual component in the second mode also manifests seasonal anomaly of the ionosphere at higher mid-latitudes around noontime. It is concluded that the EOF analysis, as a statistical eigen mode method, is resultful in analyzing the ionospheric climatology hence can be used to construct the empirical model for the ionospheric climatology.

TDS-1 GNSS reflectometry wind geophysical model function response to GPS block types

This paper presents the TDS-1 GNSS reflectometry wind Geophysical Model Function(GMF)response to GPS block types.The observables were extracted from Delay Doppler Maps(DDMs)after taking the receiver antenna gains effects and GNSS-R geometry effects into account.Since the DDM is affected by GPS EffectiveIsotropic Radiated Power(EIRP),we first investigate the sensitivity of observables to the GPS block.Additionally,the observables at high SNRs are more sensitive to wind speed,but the spatial coverage at high signal to noise ratios(SNRs)is lower,while DDMs at low SNRs have the opposite characteristics.To balance the accuracy and spatial coverage,the DDM datasets are divided into two parts:high SNR(>0 dB)and low SNR(>−10 dB and≤0 dB)to develop wind GMF.Then,the influences of GPS block on wind speed retrieval both at high and low SNR is analyzed.Results show that the block types have impacts on wind GMF and the use of a prior GPS block can contribute to a better wind speed retrieval both at high and low SNR.Compared with ASCAT,the Root Mean Square Error(RMSE)value of wind speed retrieval at high and low SNR are 2.19 m/s and 3.13 m/s,respectively,when all TDS data are processed without distinguishing GPS block types.However,if the TDS data are separately processed and used to develop wind GMF through different blocks,both the accuracy and correlation coefficient can be improved to some extent.Finally,the influence of significant height of the swell(Hs)on SNR observables is analyzed,and it is demonstrated that there is no obvious linear or nonlinear relationship between them.

GPT/GMF和GPT2w/VMF1对流层模型在我国境内的适用性分析

采用陆态网的观测数据,开展非差无电离层组合PPP实验,分析了GPT/GMF和GPT2w/VMF1对流层模型在我国境内的适用性。结果表明,在陆态网基准站坐标的水平分量上,二者的吻合性较好;但在高程分量上,二者的差异性较大,大约有33%的站坐标高程分量的差异超过10mm,这些基准站集中分布在海拔较高的青藏高原地区。采用非差模式解算我国西部海拔较高地区的GNSS测站观测数据时,建议选择最新、精度最高且完全公开的GPT2w/VMF1模型。

气候模式CAS-ESM-C对1月热带太平洋流场模态的模拟评估

利用气候模式CAS-ESM-C从1922年起84年的模拟资料,对1月份热带太平洋上层流场作复EOF分解及小波分析,并与实况以及理论解析解作对比讨论,以考察模式对赤道大洋上层流场的模拟能力,得到主要结论:(1)复EOF分解前3个模态的方差贡献为53.5%、12.9%、9.5%,累积方差贡献为75.9%,累积方差贡献比实况更高。(2)第一、二模态空间场与实况相比总体相像,流场都为赤道所俘获,在俘获区内的流场均以偏纬向流为主;差异在于模拟资料分析的赤道俘获区范围较实况要大,流场的经向流分量及越赤道流也较实况明显。(3)第一、二模态实时间系数序列无线性变化趋势,而实况则有。复EOF模态年际及年代际变化与实况相同或相近;第一、二模态中3~7年的年际变化是厄尔尼诺与南方涛动(ENSO)的反映;第一模态22~23年的年代际变化受北太平洋主要气候模态北太平洋年代际振荡(PDO)对热带太平洋的影响,而第二模态13年的年代际变化是受北太平洋次要气候模态北太平洋环流振荡(NPGO)对热带太平洋的影响;第一、二模态还都有峰值16年的年代际变化,这可能与印尼穿越流有关。(4)模拟资料分析的结果具有理论解析解中流场为赤道所俘获及流场为纬向流的特点,只是解析解中因无风应力强迫,流场呈纯纬向流。(5)第一(二)模态在赤道太平洋东部(中部)有海温动力异常,并可称之为东(中)部型ENSO模态。气候模式CAS-ESM-C对热带太平洋上层流场的模拟表现较佳。

燃料电池模型多尺度参数双代价函数的全局灵敏度分析

质子交换膜燃料电池半经验模型存在较多未知经验参数,且其尺度和取值范围均相差较大,在全局灵敏度分析中局部高灵敏度信息易丢失,从而导致结果存在偏差。为此,构造了互为倒数的双代价函数。在原代价函数基础上计算全域内经验参数与响应误差的相关性,利用Sobol方法均匀采样进行首次全局灵敏度分析,筛选整个取值范围内高灵敏度(即影响收敛速度)参数;进而利用倒数代价函数放大局部区域的误差相关性,对于剩余全域不敏感参数,进行倒数代价函数的再次全局灵敏度分析,筛选出局部取值范围内高灵敏度(即影响收敛精度)参数。从而提高了多尺度多局域高灵敏参数识别能力。分析结果证明了所得高灵敏度参数辨识后模型的响应误差与全参数辨识的结果一致,可节省约60%的计算成本。该方法的适用性和准确性在电堆实验中均得到了有效验证。

EMD-LSTM-LB分频时序预测算法

针对现有时间序列预测模型存在的误差大、不稳定等问题,提出一种基于EMD的时间序列预测模型。该模型适应于预测数据量大、波动性大的时间序列,提出一种分配函数,解决分解后子序列的分配问题。EMD将原始序列分解为一系列子序列,分配函数根据子序列的波动状态划分出极端子序列和非极端子序列。极端子序列由LSTM进行训练预测,非极端子序列子序列由LB进行训练预测,将所有子序列整合为最终预测结果。与已有的一些模型方法相比,各数据集的预测精度均有所提高。

基于统计感知策略的高斯混合模型求解方法

高斯混合模型(Gaussian mixture model,GMM)是一种经典的概率模型,常被用于无监督学习领域来确定无类别标记样本点的类别分布。作为求解GMM参数的重要技术,期望最大化(Expectation maximization,EM)算法通过计算GMM对应似然函数的最优解确定基模型自身参数以及基模型的混合系数。利用EM算法求解GMM存在如下两个缺陷:EM算法易于陷入局部最优解以及EM算法确定GMM基模型相关参数的不稳定,尤其是针对多维随机变量。本文提出了一种基于统计感知(Statistical⁃aware,SA)策略的GMM求解方法——SA⁃GMM方法。该方法从估计给定数据集的未知概率密度函数入手,建立了核密度估计(Kernel density estimation,KDE)与GMM之间的关联。为避免KDE对“过平滑”窗口的选取,设计了同时最小化KDE与GMM之间的经验风险和KDE窗口结构风险的目标函数,进而确定了GMM的最优参数。在11个标准概率分布上的实验证明了SA⁃GMM方法的可行性、合理性和有效性,同时结果也表明SA⁃GMM能够获得显著优于基于EM算法的GMM及其变体的概率密度函数估计表现。

我国地震人员损失评估经验概率模型

对地震人员伤亡的准确预测是震后救灾资源调度和救援人员分配的重要依据。基于PAGER系统中的双参数累积对数正态分布模型,分别讨论了3种建立我国地震人员损失预测模型的方法,并通过历史地震的受灾情况对其预测效果进行验证。其中,直接采用PAGER系统中提供的我国参数和以极震区烈度为变量建立的预测模型的准确性均不理想。考虑到多数震害调查资料中各烈度区受灾数据的缺失,选取数据较为详实的2008年四川汶川8.0级地震,根据已知受地震影响的25个市县的人员损失总数,综合考虑烈度区面积、人口数和烈度对地震人员损失数的影响,将每个市县的人员损失总数进行分配,最终得到不同烈度区的人员死亡率和受灾人数,回归得到模型参数后,建立我国地震人员损失评估经验概率模型。经验证分析,该模型预测结果与震例结果十分接近,远好于前2种方法得到的预测模型。同时,该模型的残差频率的正态性较好,根据预测结果建议的地震应急响应级别均与震例的实际情况相符。

基于时空分解的汾河流域植被净初级生产力的变异性及驱动因子分析

汾河流域生态环境脆弱,且水土流失问题较为严重,易受自然及人为因素的影响。基于1982—2019年汾河流域的植被净初级生产力(NPP),利用趋势分析、经验正交分解(EOF)和结构方程模型(SEM)研究汾河流域NPP的时空演变特征及其驱动因素,旨在为汾河流域生态环境建设和保护提供理论依据。结果表明,汾河流域NPP由流域中心向边缘增加,整体呈波动上升趋势,NPP显著增加的面积远大于退化面积;从2003年开始,汾河流域尤其是东北部地区植被NPP明显提高,中北部区域的NPP呈现退化趋势,并表现出较强的时间变异性。人类活动、气候变化和基础自然环境因子对汾河流域NPP空间结构的总效应分别为0.50、0.20、0.03,表明了人类活动是影响NPP空间结构最主要的因素;基础自然环境对NPP变化有显著的直接和间接效应,并且容易在小尺度范围内形成潜在的时空格局。温度和相对湿度是影响流域NPP空间结构和时间变异最主要的气候因素。