A Gaussian pseudolikelihood approach for quantile regression with repeated measurements

To enhance the efficiency of regression parameter estimation by modeling the correlation structure of correlated binary error terms in quantile regression with repeated measurements, we propose a Gaussian pseudolikelihood approach for estimating correlation parameters and selecting the most appropriate working correlation matrix simultaneously. The induced smoothing method is applied to estimate the covariance of the regression parameter estimates, which can bypass density estimation of the errors. Extensive numerical studies indicate that the proposed method performs well in selecting an accurate correlation structure and improving regression parameter estimation efficiency. The proposed method is further illustrated by analyzing a dental dataset.     

Reweighting anthropometric data using a nearest neighbour approach

When designing products and environments, detailed data on body size and shape are seldom available for the specific user population. One way to mitigate this issue is to reweight available data such that they provide an accurate estimate of the target population of interest. This is done by assigning a statistical weight to each individual in the reference data, increasing or decreasing their influence on statistical models of the whole. This paper presents a new approach to reweighting these data. Instead of stratified sampling, the proposed method uses a clustering algorithm to identify relationships between the detailed and reference populations using their height, mass, and body mass index (BMI). The newly weighted data are shown to provide more accurate estimates than traditional approaches. The improved accuracy that accompanies this method provides designers with an alternative to data synthesis techniques as they seek appropriate data to guide their design practice.Practitioner Summary: Design practice is best guided by data on body size and shape that accurately represents the target user population. This research presents an alternative to data synthesis (e.g. regression or proportionality constants) for adapting data from one population for use in modelling another.     

A cost model for spatio-temporal queries using the TPR-tree

A query optimizer requires cost models to calculate the costs of various access plans for a query. An effective method to estimate the number of disk (or page) accesses for spatio-temporal queries has not yet been proposed. The TPR-tree is an efficient index that supports spatio-temporal queries for moving objects. Existing cost models for the spatial index such as the R-tree do not accurately estimate the number of disk accesses for spatio-temporal queries using the TPR-tree, because they do not consider the future locations of moving objects, which change continuously as time passes.In this paper, we propose an efficient cost model for spatio-temporal queries to solve this problem. We present analytical formulas which accurately calculate the number of disk accesses for spatio-temporal queries. Extensive experimental results show that our proposed method accurately estimates the number of disk accesses over various queries to spatio-temporal data combining real-life spatial data and synthetic temporal data. To evaluate the effectiveness of our method, we compared our spatio-temporal cost model (STCM) with an existing spatial cost model (SCM). The application of the existing SCM has the average error ratio from 52% to 93%, whereas our STCM has the average error ratio from 11% to 32%.     

一种新型自适应神经网络回归估计算法

结合自适应谐振理论和域理论的优点,针对回归估计问题的特性,提出了一种新型神经网络回归估计算法ＦＴＡＲＴ３．该算法学习速度快、归纳能力强,不仅具有增量学习能力,还克服了ＢＰ类算法需要人为设置隐层神经元的缺陷,直线、下弦、二维墨西哥草帽、三维墨西哥草帽等４个实验表明,ＦＴＡＲＴ３在函数近似效果和训练时间代价上都优于目前常用于因归估计问题的ＢＰ类算法。     

基于二阶统计量的近场源四维参数联合估计

研究了一种基于二阶统计量的近场窄带信源频率、方位角、仰角和距离四维参数的联合估计算法。该算法利用中心对称的十字阵列,根据阵元输出构造二阶统计量矩阵,并结合这些矩阵的特点构造新矩阵,利用特征值和特征向量信息得到信源参数。该方法仅需计算6个二阶统计量矩阵就可以估计四维参数,此外参数自动配对。仿真结果表明该算法能准确地估计近场源参数。     

Determining Sample Characteristics and Their Asymptotic Linear-Regression Properties Estimated Using Inequality Constraints

Linear regression parameters are estimated using inequality constraints. A sample estimate of the noise variance is proposed and its consistency is proved. A sample estimate of the matrix of mean-square errors in estimates of regression parameters is considered. Its consistency is proved for rather general assumptions on the noise distribution law. __________ Translated from Kibernetika i Sistemnyi Analiz, No. 3, pp. 153–165, May–June 2005.     

A factor-based probabilistic cost model to support bid-price estimation

An appropriate bid price is essential for winning a construction project contract. However, making an accurate cost estimate is both time-consuming and expensive. Thus, a method that does not take much time and can approximate a proper bid price can help a contractor in making bid-price decisions when the available bid-estimation time is insufficient. Such a method can also generate a target cost and provide a cross-check for their bid prices that were estimated using a detailed process. This study proposes a novel model for quickly making a bid-price estimation that integrates a probabilistic cost sub-model and a multi-factor evaluation sub-model. The cost sub-model, which is simulation-based, focuses on the cost divisions to save estimation time. At the same time, the multi-factor evaluation sub-model captures the specific factors affecting the cost of each cost division. The advantages of the proposed model are demonstrated by its application to three residential housing projects located in northern Taiwan. The steps for applying this model to other contractors are also provided.     

一种小规模数据集下的贝叶斯网络学习方法及其应用

提出了一种小规模数据集下学习贝叶斯网络的有效算法—FCLBNo FCLBN利用bootstrap方法在给定的小样本数据集上进行重抽样,然后用在抽样后数据集上学到的贝叶斯网络来佑计原数据集上的贝叶斯网络的高置信度的特征,并用这些特征来指导在原数据集上的贝叶斯网络搜索。用标准的数据集验证了FCLBN的有效性,并将FCLBN应用于酵母菌细胞中蛋白质的定位预测。实验结果表明,FCLBN能够在小规模数据集上学到较好的网络模型。     

Estimating sparse spectro-temporal receptive fields with natural stimuli

Several algorithms have been proposed to characterize the spectro-temporal tuning properties of auditory neurons during the presentation of natural stimuli. Algorithms designed to work at realistic signal-to-noise levels must make some prior assumptions about tuning in order to produce accurate fits, and these priors can introduce bias into estimates of tuning. We compare a new, computationally efficient algorithm for estimating tuning properties, boosting, to a more commonly used algorithm, normalized reverse correlation. These algorithms employ the same functional model and cost function, differing only in their priors. We use both algorithms to estimate spectro-temporal tuning properties of neurons in primary auditory cortex during the presentation of continuous human speech. Models estimated using either algorithm, have similar predictive power, although fits by boosting are slightly more accurate. More strikingly, neurons characterized with boosting appear tuned to narrower spectral bandwidths and higher temporal modulation rates than when characterized with normalized reverse correlation. These differences have little impact on responses to speech, which is spectrally broadband and modulated at low rates. However, we find that models estimated by boosting also predict responses to non-speech stimuli more accurately. These findings highlight the crucial role of priors in characterizing neuronal response properties with natural stimuli.     

Software development cost estimation using wavelet neural networks

Software development has become an essential investment for many organizations. Software engineering practitioners have become more and more concerned about accurately predicting the cost and quality of software product under development. Accurate estimates are desired but no model has proved to be successful at effectively and consistently predicting software development cost. In this paper, we propose the use of wavelet neural network (WNN) to forecast the software development effort. We used two types of WNN with Morlet function and Gaussian function as transfer function and also proposed threshold acceptance training algorithm for wavelet neural network (TAWNN). The effectiveness of the WNN variants is compared with other techniques such as multilayer perceptron (MLP), radial basis function network (RBFN), multiple linear regression (MLR), dynamic evolving neuro-fuzzy inference system (DENFIS) and support vector machine (SVM) in terms of the error measure which is mean magnitude relative error (MMRE) obtained on Canadian financial (CF) dataset and IBM data processing services (IBMDPS) dataset. Based on the experiments conducted, it is observed that the WNN-Morlet for CF dataset and WNN-Gaussian for IBMDPS outperformed all the other techniques. Also, TAWNN outperformed all other techniques except WNN.     

Consumer credit risk: Individual probability estimates using machine learning

Consumer credit scoring is often considered a classification task where clients receive either a good or a bad credit status. Default probabilities provide more detailed information about the creditworthiness of consumers, and they are usually estimated by logistic regression. Here, we present a general framework for estimating individual consumer credit risks by use of machine learning methods. Since a probability is an expected value, all nonparametric regression approaches which are consistent for the mean are consistent for the probability estimation problem. Among others, random forests (RF), k-nearest neighbors (kNN), and bagged k-nearest neighbors (bNN) belong to this class of consistent nonparametric regression approaches. We apply the machine learning methods and an optimized logistic regression to a large dataset of complete payment histories of short-termed installment credits. We demonstrate probability estimation in Random Jungle, an RF package written in C++ with a generalized framework for fast tree growing, probability estimation, and classification. We also describe an algorithm for tuning the terminal node size for probability estimation. We demonstrate that regression RF outperforms the optimized logistic regression model, kNN, and bNN on the test data of the short-term installment credits.     

Neural network cost estimates for heart bypass surgery

This paper reports on the results of using artificial neural network (ANN) technology to estimate treatment costs of heart bypass patients based on their diagnostic condition and clinical criteria. Our applications include: (1) predicting total episode cost using clinical data; (2) a method for providing rapid feedback to assess change in total costs within a turbulent environment; and (3) a procedure for identifying activity-based cost driver candidates that would normally not surface from an analysis of accounting data. Clinical data were collected on 250 heart bypass patients at the University of Ottawa Heart Institute. The data analysed support the following conclusions: (1) clinical and diagnostic indicators obtained before surgery for individual patients can be used to estimate the total cost of their heart bypass surgery; (2) the average cost estimate error decreases as we add clinical information available during and after the surgical event; (3) the procedure used to estimate individual patient cost does not require access to accounting records; (4) the forecasting system we describe may improve exception reporting for individual patients by tracking costs and clinical criteria on a real-time basis throughout the treatment episode.     

相位调制信号的小波参数估计方法及其Cramer-Rao界

现代雷达广泛使用相位调制信号,相位参数的有效估计对辐射源的识别和分类具有重要意义。运用小波变换方法对多项式相位信号的相位系数进行了估计,给出了小波变换方法下的估计多项式相位调制信号相位系数的Cramer-Rao下限,并提出了一种新的相位系数估计算法。理论分析和仿真结果表明利用小波变换方法估计相位调制信号的相位系数参数估值更精确,可以得到更低的Cramer-Rao下限。     

The Performance of German Firms in the Business-Related Service Sectors Revisited: Differential Evolution Markov Chain Estimation of the Multinomial Probit Model

We outline a new estimation method for the multinomial probit model (MNP). The method is a differential evolution Markov chain algorithm that employs a Metropolis-within-Gibbs sampler with data augmentation and the Geweke–Hajivassiliou–Keane (GHK) probability simulator. The method lifts the curse of dimensionality in numerical integration as it neither requires simulation of the whole likelihood function nor the computation of its analytical or numerical derivatives. The method is applied to an unbalanced panel dataset of firms from the German business-related service sector over the period 1994–2000. In spite of its less restricted character, the MNP model is found not to provide more accurate estimates for explaining the performance of these firms than the multinomial logit model.     

Cost-sensitive and modular land-cover classification based on posterior probability estimates

Many types of nonlinear classifiers have been proposed to automatically generate land-cover maps from satellite images. Some are based on the estimation of posterior class probabilities, whereas others estimate the decision boundary directly. In this paper, we propose a modular design able to focus the learning process on the decision boundary by using posterior probability estimates. To do so, we use a self-configuring architecture that incorporates specialized modules to deal with conflicting classes, and we apply a learning algorithm that focuses learning on the posterior probability regions that are critical for the performance of the decision problem stated by the user-defined misclassification costs. Moreover, we show that by filtering the posterior probability map, the impulsive noise, which is a common effect in automatic land-cover classification, can be significantly reduced. Experimental results show the effectiveness of the proposed solutions on real multi- and hyperspectral images, versus other typical approaches, that are not based on probability estimates, such as Support Vector Machines.     

A hybrid cost estimation framework based on feature-oriented data mining approach

This paper presents an informatics framework to apply feature-based engineering concept for cost estimation supported with data mining algorithms. The purpose of this research work is to provide a practical procedure for more accurate cost estimation by using the commonly available manufacturing process data associated with ERP systems. The proposed method combines linear regression and data-mining techniques, leverages the unique strengths of the both, and creates a mechanism to discover cost features. The final estimation function takes the user’s confidence level over each member technique into consideration such that the application of the method can phase in gradually in reality by building up the data mining capability. A case study demonstrates the proposed framework and compares the results from empirical cost prediction and data mining. The case study results indicate that the combined method is flexible and promising for determining the costs of the example welding features. With the result comparison between the empirical prediction and five different data mining algorithms, the ANN algorithm shows to be the most accurate for welding operations.     

A robust multimedia surveillance system for people counting

Closed circuit television cameras (CCTV) are widely used in monitoring. This paper presents an intelligent CCTV crowd counting system based on two algorithms that estimate the density of each pixel in each frame and use it as a basis for counting people. One algorithm uses scale-invariant feature transform (SIFT) features and clustering to represent pixels of frames (SIFT algorithm) and the other uses features from accelerated segment test (FAST) corner points with SIFT features (SIFT-FAST algorithm). Each algorithm is designed using a novel combination of pixel-wise, motion-region, grid map, background segmentation using Gaussian mixture model (GMM) and edge detection. A fusion technique is proposed and used to validate the accuracy by combining the result of the algorithms at frame level. The proposed system is more practical than the state of the art regression methods because it is trained with a small number of frames so it is relatively easy to deploy. In addition, it reduces the training error, set-up time, cost and open the door to develop more accurate people detection methods. The University of California (UCSD) and Mall datasets have been used to test the proposed algorithms. The mean deviation error, mean squared error and the mean absolute error of the proposed system are less than 0.1, 16.5 and 3.1, respectively, for the Mall dataset and less than 0.07, 5.5 and 1.9, respectively, for UCSD dataset.

     

基于遗传算法和案例推理的软件费用估算方法

为了提高类比法的估算精度,减少人工检索案例的工作量和难度,提出了一种基于遗传算法和案例推理的软件费用估算方法。给出了案例推理过程的估算步骤,构造了案例的相似性度量函数;设计了用于案例推理问题的遗传算法,利用该算法在历史数据库中搜索与目标案例最相似的项目,并对软件项目的特征权重进行优化;借助Albrecht和Desharnais数据库,对提出的方法进行分析。实验结果表明,该方法可以在软件生命周期的早期显著提高软件费用的估算精度,与类比估算和线性回归方法相比,具有更小的平均误差率,其估算性能符合相关行业软件的估算需求。     

DiCoMo: the digitization cost model

The estimate of digitization costs is a very difficult task. It is difficult to obtain accurate values because of the great quantity of unknown factors. However, digitization projects need to have a precise idea of the economic costs and the times involved in the development of their contents. The common practice when we start digitizing a new collection is to set a schedule, and a firm commitment to fulfil it (both in terms of cost and deadlines), even before the actual digitization work starts. As it happens with software development projects, incorrect estimates produce delays and cause costs overdrafts. Based on methods used in Software Engineering for software development cost prediction like COCOMO and Function Points, and using historical data gathered during 5 years at the MCDL project, during the digitization of more than 12000 books, we have developed a method for time-and-cost estimates named DiCoMo (Digitization Cost Model) for digital content production in general. This method can be adapted to different production processes, like the production of digital XML or HTML texts using scanning and OCR, and undergoing human proofreading and error correction, or for the production of digital facsimiles (scanning without OCR). The accuracy of the estimates improve with time, since the algorithms can be optimized by making adjustments based on historical data gathered from previous tasks. Finally, we consider the problem of parallelizing tasks, i.e. dividing the work among a number of encoders that will work in parallel.     

A generalized regression-based model for forecasting winter wheat yields in Kansas and Ukraine using MODIS data

Wheat is one of the key cereal crops grown worldwide, providing the primary caloric and nutritional source for millions of people around the world. In order to ensure food security and sound, actionable mitigation strategies and policies for management of food shortages, timely and accurate estimates of global crop production are essential. This study combines a new BRDF-corrected, daily surface reflectance dataset developed from NASA's Moderate resolution Imaging Spectro-radiometer (MODIS) with detailed official crop statistics to develop an empirical, generalized approach to forecast wheat yields. The first step of this study was to develop and evaluate a regression-based model for forecasting winter wheat production in Kansas. This regression-based model was then directly applied to forecast winter wheat production in Ukraine. The forecasts of production in Kansas closely matched the USDA/NASS reported numbers with a 7% error. The same regression model forecast winter wheat production in Ukraine within 10% of the official reported production numbers six weeks prior to harvest. Using new data from MODIS, this method is simple, has limited data requirements, and can provide an indication of winter wheat production shortfalls and surplus prior to harvest in regions where minimal ground data is available.