A study on the persistence of Farrell’s efficiency measure under a dynamic framework

This article introduces a sequence of four systematic methods to examine the extent to which the economic efficiency of Taiwan’s commercial banks persists and to uncover the potential dynamic link between bank performance and various financial indicators. Quasi-fixed inputs are explicitly incorporated in the DEA model to account for possible adjustment costs, regulation, or indivisibilities. Among the four methods, the dynamic panel data model and the Markov model appear to be exploited for the first time in the area of the DEA approach. Evidence is found that bank efficiency exhibits moderate persistence over the sample period, implying that the given sample banks fail to adjust their production techniques in a timely manner. Regulatory authorities and bank managers are suggested to be aware of the level of undesirable non-performing loans due to their close relationship with bank performance.     

一种基于特征提取技术的语音识别算法

语音识别是通过对说话人语音信号的分析和提取．自动确定说话人是否在所登记的说话人集合中，以及说话的人是谁的过程．本文提出了一种基于特征提取技术的模板匹配语音识别算法．该算法中，采用与文本有关（主要用于时域分析）和与文本无关（主要用于频域分析）相结合的三种识别方法：快速傅立叶变换（FFT）、平均过零率分析（RBZ）和高斯混合模型（GMM），最后通过三种方法的加权平均来生成判决门限值．通过比较门限值误差的大小，根据“最大似然准则”给出待识别语音的身份．     

成交量与价格限制对中国股市影响的实证分析

以我国股票市场的12支个股资料为例,研究了日收益相关性、成交量和价格限制的关系.研究发现,增加成交量会减弱自相关性的假设对近一半的股票成立.我们首先采用一般方法来估计模型,但这种方法用于研究这三者的关系存在一定的偏差,所以接下来又采用了广义矩估计(GMM)来观察估计结果的灵敏性.发现用GMM方法得出的结果表明成交量和价格限制的影响要强于一般方法得出的结果,价格限制影响表现出了对自回归系数更强烈的影响.     

基于EM和GMM相结合的自适应灰度图像分割算法

提出一种阈值自适应、EM方法估计GMM参量的图像分割算法,能够根据图像的内容结合区域和边界两方面的信息自适应地选择阈值,精确地进行图像边界分割.算法首先提取图像的边界,然后根据边界的直方图计算图像的可分割性,由可分割性确定EM方法的阈值进行GMM分割,最后合并图像的近似区域.实验数据表明,相比其它图像分割算法,以及固定阈值的传统EM算法,本算法的分割结果更为准确.     

基于改进高斯混合模型的运动目标检测与跟踪算法

针对传统高斯模型存在的不足,为了搞高运动目标跟踪精度,提出一种基于改进高斯混合模型的目标检测与跟踪算法。首先提取目标特征建立目标分类器,并将目标从前景是标记出来;然后通过多目标跟踪将目标为多种运动模式;最后采用高斯混合模型对跟踪与分类的结果进行融合,获得最终目标的位置。结果表明,本文方法不仅提高了目标检测与跟踪精度,而且可高斯模型以较好的满足目标跟踪的实时性要求。     

Hybrid segmentation of left ventricle in cardiac MRI using gaussian-mixture model and region restricted dynamic programming

Segmentation of the left ventricle from cardiac magnetic resonance images (MRI) is very important to quantitatively analyze global and regional cardiac function. The aim of this study is to develop a novel and robust algorithm which can improve the accuracy of automatic left ventricle segmentation on short-axis cardiac MRI. The database used in this study consists of three data sets obtained from the Sunnybrook Health Sciences Centre. Each data set contains 15 cases (4 ischemic heart failures, 4 non-ischemic heart failures, 4 left ventricle (LV) hypertrophies and 3 normal cases). Three key techniques are developed in this segmentation algorithm: (1) ray scanning approach is designed for segmentation of images with left ventricular outflow tract (LVOT), (2) a region restricted technique is employed for epicardial contour extraction, and (3) an edge map with non-maxima gradient suppression approach is put forward to improve the dynamic programming to derive the epicardial boundary. The validation experiments were performed on a pool of data sets of 45 cases. For both endo- and epi-cardial contours of our results, percentage of good contours is about 91%, the average perpendicular distance is about 2 mm. The overlapping dice metric is about 0.92. The regression and determination coefficient between the experts and our proposed method on the ejection fraction (EF) is 1.01 and 0.9375, respectively; they are 0.9 and 0.8245 for LV mass. The proposed segmentation method shows the better performance and is very promising in improving the accuracy of computer-aided diagnosis systems in cardiovascular diseases.     

True and apparent scaling: The proximity of the Markov-switching multifractal model to long-range dependence

In this paper, we consider daily financial data of a collection of different stock market indices, exchange rates, and interest rates, and we analyze their multi-scaling properties by estimating a simple specification of the Markov-switching multifractal (MSM) model. In order to see how well the estimated model captures the temporal dependence of the data, we estimate and compare the scaling exponents H(q) (for q=1,2) for both empirical data and simulated data of the MSM model. In most cases the multifractal model appears to generate ‘apparent’ long memory in agreement with the empirical scaling laws.