基于测井资料的低孔低渗储层产能预测研究

针对低孔低渗储层产能预测模型受敏感因素影响大，主次因素在低产情况下区别不明显，产能预测评价相对较困难等实际问题，以准噶尔盆地夏子街地区夏77井、夏79井为例，开展低孔低渗储层产能预测的前期试验研究，对多个产能预测模型及其适用条件进行了对比分析，从Darcy模型、Jones（1976）模型、Vogel/Harrison（1968）模型、裂缝模型中进行优选和调试。通过分析产能预测模型结果与测试结果在产量上的相关性，表明基于测井资料的预测结果与试油测试结果相当吻合，从而确认产能预测模型结果的可靠性。     

The weighted jackknife for ratio estimation

Using the idea of impirical influence function, Hinkley (1977), the weighted jackknife technique is extended to ratio estimation. A weighted jackknife variance estimator for the ratio estimator is developed. Using the prediction theory approach, the properties of the weighted jackknifed variance estimator are examined. The implications of the failures of regression model on the behaviour of the weighted jackknifed variance estimator, for ratio estimation, are also studied.     

Local predictive influence in bayesian linear models with conjugate priors

Cook (1986) presented the idea of local influence to study the sensitivity of inferences to model assumptions:introduce a vector δ of perturbations to the model; choose a discrepancy function D to measure differences between the original inference and the inference under the perturbed model; study the behavior of D near δ = 0, the original model, usually by taking derivatives. Johnson and Geisser (1983) measure influence in Bayesian inference by the Kullback-Leibler divergence between predictive distributions. I~IcCulloch (1989) is a synthesis of Cook and Johnson and Geisser, using Kullback-Leibler divergence between posterior or predictive distributions as the discrepancy function in Bayesian local influence analyses. We analyze a special case for which McCulloch gives the general theory; namely, the linear model with conjugate prior. We present specific formulae for local influence measures for 1) changes in the parameters of the gamma prior for the precision, 2) changes in the mean of the normal prior for the regression coefficients, 3) changes in the covariance matrix of the normal prior for the regression coefficients and 4) changes in the case weights. Our method is an easy way to find locally influential subsets of points without knowing in advance the sizes of the subsets. The techniques are illustrated with a regression example.     

Ecological Risk Assessment,Prediction, and Assessing Risk Predictions

Ecological risk assessment embodied in an adaptive management framework is becoming the global standard approach for formally assessing and managing the ecological risks of technology and development. Ensuring the continual improvement of ecological risk assessment approaches is partly achieved through the dissemination of not only the types of risk assessment approaches used, but also their efficacy. While there is an increasing body of literature describing the results of general comparisons between alternate risk assessment methods and models, there is a paucity of literature that post hoc assesses the performance of specific predictions based on an assessment of risk and the effectiveness of the particular model used to predict the risk. This is especially the case where risk assessments have been used to grant consent or approval for the construction of major infrastructure projects. While postconstruction environmental monitoring is increasingly commonplace, it is not common for a postconstruction assessment of the accuracy and performance of the ecological risk assessment and underpinning model to be undertaken. Without this “assessment of the assessment,” it is difficult for other practitioners to gain insight into the performance of the approach and models used and therefore, as argued here, this limits the rate of improvement of risk assessment approaches.     

On the Nonlinear Predictability of Stock Returns Using Financial and Economic Variables

In a recent article by Qi, neural networks trained by Bayesian regularization were used to predict excess returns on the S&P 500. The article concluded that the switching portfolio based on the recursive neural-network forecasts generates higher accumulated wealth with lower risks than that based on linear regression. Unfortunately, attempts to replicate the results were unsuccessful. Replicated results using the same software, approach and data detailed by Qi indicate that, in fact, the switching portfolio based on the recursive neural-network forecasts generates lower accumulated wealth with higher risks than that based on linear regression.     

基于贝叶斯正则化的TDBPNN模型在中国外贸预报中的应用及评估

基于相空间重构的非线性预报思想,建立一个时滞的BP神经网络模型(TDBPNN),采用贝叶斯正则化方法提高BP网络的泛化能力,并将该模型应用于中国进出口贸易的预测,结果证明改进的TDBPNN模型具有较好的泛化能力,准确拟合了进出口贸易发展的历史值和趋势。区别于一般的预测评价,认为非线性预测不仅要注重数据拟合和精度改进,而且应该能够反映被预报系统的非线性特征。在分析模型预测精度的同时,通过计算拟合序列和原序列的非线性特征量进行模型评价,证实预测模型能够合理地“捕捉”到产生原序列的非线性系统的动力学特征。     

Reliable Ad Hoc Routing Based on Mobility Prediction

Reliability is a very important issue in Mobile Ad hoc NETworks (MANETs). Shortest paths are usually used to route packets in MANETs. However, a shortest path may fail quickly, because some of the wireless links along a shortest path may be broken shortly after the path is established due to mobility of mobile nodes. Rediscovering routes may result in substantial data loss and message exchange overhead. In this paper, we study reliable ad hoc routing in the urban environment. Specifically, we formulate and study two optimization problems. In the minimum Cost Duration-bounded Path (CDP) routing problem, we seek a minimum cost source to destination path with duration no less than a given threshold. In the maximum Duration Cost-bounded Path (DCP) routing problem, we seek a maximum duration source to destination path with cost no greater than a given threshold. We use a waypoint graph to model the working area of a MANET and present an offline algorithm to compute a duration prediction table for the given waypoint graph. An entry in the duration prediction table contains the guaranteed worst-case duration of the corresponding wireless link. We then present an efficient algorithm which computes a minimum cost duration-bounded path, using the information provided in the duration prediction table. We also present a heuristic algorithm for the DCP routing problem. In addition, we show that the proposed prediction and routing schemes can be easily applied for designing reliable ad hoc routing protocols. Simulation results show that our mobility prediction based routing algorithms lead to higher network throughput and longer average path duration, compared with the shortest path routing. This research was supported in part by ARO grant W911NF-04-1-0385 and NSF grant CCF-0431167. The information reported here does not reflect the position or the policy of the federal government.     

企业经济活动预测问题的一种数学模型及其应用

介绍了企业经济活动预测问题的一种数学模型 ,通过矩阵的运算 ,运用Matlab数学软件 ,使问题迅速得到解决 ,并通过举例说明了其应用     

基于LSTM神经网络的金融时间序列预测

本文提出将小波分析与纳入时间序列依赖特征的长短期记忆（LSTM）神经网络相结合，构建金融时间序列数据预测模型，以克服现有模型对金融时间序列数据非平稳、非线性、序列相关等复杂特征以及数据间非线性交互关系无法反映的缺陷。同时，以道琼斯工业指数日收盘价为例，探究LSTM神经网络对实际金融时间序列数据的预测能力，比较其与多层感知机、支持向量机、K近邻、GARCH四种模型的预测效果。实证结果表明LSTM神经网络具有更高的预测精度，能够有效预测金融时间序列数据的长短期动态变化趋势，说明了其对金融时间序列数据预测的适用性与有效性。此外，对金融时间序列数据进行小波分解与重构，可有效提高LSTM预测模型的泛化能力，以及对长短期动态趋势的预测精度。     

A correlation-theoretical interpretation of schur analysis

This paper is heavily leaned on the author's recent investigations concerning SCHUR analysis of non-negative Hermitian block matrices. The parameters of the matrix balls and the triangular choice scheme which describe a non-negative Hermitian block matrix will be interpreted in the framework of correlation theory