Recent Activities

Seminar: A Note on Bartlett's M Test for Homogeneity of Variances
Speaker: Dr. Lingyun (Larry) Zhang, United International College
Time: 3.30pm-4.00pm, Oct 19 (Wednesday), 2011
Absract:
After pointing out a drawback in Bartlett?ˉs chi-square approximation, we suggest a simple modification and a Gamma approximation to improve Bartlett?ˉs M test for homogeneity of variances. 
Seminar: Optimal Dividend Distribution under Markov-regime Switching
Speaker: Dr. Jiang Zhengjun, United International College
Time: 2:00 pm - 3:00 pm, Dec 8 (Wednesday), 2010
Absract:
We consider optimal dividend distribution for a company whose cumulative net revenues evolve as a Markov-regime switching Brownian motion with drift and discounting rate. The goal of the company is to maximize the expected cumulative discounted dividends until the first time that cash reserve (cumulative net revenues minus cumulative dividends) is zero. For positive drift in each regime, a regime-dependent barrier strategy is optimal and value function is the fixed point of a contraction. For negative drift in some regimes, optimal dividend policy is different. 
Seminar: Statistical Applications to Financial Risk Management
Speaker: Dr. Aijun Zhang, Bank of America
Time: 2:00 pm, Wed, June 30, 2010
Absract:
Statistical theory and methods are widely used in banking and finance. Based on my experiences in Bank of America, I would share two kinds of statistical developments for (a) financial crime detection and (b) post-crisis credit risk modeling. I would also share some visualization examples that make no more boring data. 
Seminar: Quadratic Forms in Statistics
Speaker: Prof. Kai-Tai Fang, United International College
Time: 9:00 am, Wed, June 30, 2010
Absract:
Quadratic forms have played as important role in statistics, especially in analysis of variance, regression analysis, multivariate analysis and data mining. How to find distributions of various quadratic forms and to show independence among quadratic forms are very important in theory of statistics. In this talk I introduce some basic knowledge of the quadratic forms and some important/popular theorems, like Cochran's theorem and Craig's theorem. 
Seminar: Basins of Attraction in a Cournot Duopoly Map: Analysis, Linear Algebra, and Economics
Speaker: Prof. Douglas R. Anderson , United International College
Time: 9:00 am, Wed, June 23, 2010
Absract:
We will introduce the economic concept of a Cournot Duopoly, first introduced by French economist Cournot in 1838. Then we will analyze a nonlinear discrete Cournot duopoly introduced by Austrian economist Koppel in 1996. In particular, we will use analysis methods from difference equations and linear algebra to understand the behavior of the system in relation to certain fixed points of the system. Using the professional software package Mathematica 7.0, we will illustrate graphically the effects that various eigenvalues of the linearization have on the system. 
Seminar: DYNAMIC TRAFFIC INFORMATION RESEARCH BASED ON DATA MINING OF FLOATING CARS
Speaker: Mr Wang, Rong-Jie, United International College
Time: 2:00-3:00 pm, Wed, Apr. 21, 2009
Absract:
The research about vehicle data based on floating point in real-time traffic information has just started in China in the last two years, navigation products with dynamic real-time traffic information have appeared on the market w in China, but some products with dynamic real-time traffic information is not accurate , and traffic information didn't timely update and the information on a low coverage. In response to solve the issues above, we build traffic analysis model based on the floating car datas and achieve dynamic real-time traffic information update and provide owners for dynamic real-time traffic information service at Shenzhen. 
Seminar: An Introduction to Numerical Integration and Its Application
Speaker: Dr Li, Xian-Juan, United International College
Time: 3:00-4:00 pm, Wed, Apr. 14, 2009
Absract:
The talk give a brief introduction to numerical integration and its application. First, we describe two main kinds of methods, Newton Cotes and Gauss type quadrature, by presenting the method??s construction and some key features theoretically. Second, by laying their application on the solution of Voterra equations, and analyzing the precision, stability, computation cost, and applicability, we catch the practical value of such methods. Finally, we try to propose a new method to solve long time Voterra equations.
Seminar: Young's Integral Inequality with Upper and lower Bounds
Speaker: Dr. Robert DOUGLAS, United International College
Time: 3:00-4:00 pm, Wed, Nov 31, 2009
Absract:
Young's integral inequality is reformulated with upper and lower bounds for the remainder. The new inequalities improve Young's integral inequality on all time scales, such that the case where equality holds becomes particularly transparent in this new presentation. The corresponding results for difference equations are given, and several examples are included. We extend these results to piecewise-monotone functions as well.
Seminar: Technique of Inverse Substructuring Dynamic Analysis for Complex Mechanical Structures
Speaker: Dr. Lu, Guang Qing, United International College
Time: 3:00-4:00 pm, Wed, Nov 25, 2009
Absract:
Dynamic analysis is always a focus in the design and manufacturring of mechanical and or electric products with vibratory responses and noises, especially for the ones with complex mechanical structures. Conventional theories and their applications, such as modal analysis, FEM, and statistical energy or power flow approaches, etc., estimate system dynamic responses (vibration & noise) due to excitations (dynamic forces) in a positive direction from component (substructures ) level to system (products) level. In the project to be presented, a new method C Inverse Substructuring Dynamic Analysis approach is established. It has much less the shortcomings existing in the convential theories. In the study, a unified theory for coupling structural-acoustic components derived from different sources is formulated on the basis of a two-level substructures. It applies the inverse scheme to solving vibration and noise problem. The developed set of formula can be widely used to determine either individual component frequency response function (FRF) and systmatic responses and trouble-shooting on dynamic quality of products assembly. The inverse scheme is verified via either lumped mass model or experimental study on the products of real vehicles. At the end of this presentation, some current works will be also introduced.
Seminar: A Space-Time Spectral Method for the Time Fractional Diffusion Equation
Speaker: Dr. Li, Xian Juan, United International College
Time: 3:00-4:00 pm, Wed, Nov 18, 2009
Absract:
In this paper, our work is focused on the theoretical investigation and numerical computation of the fractional diffusion equations (FDEs), which are of interest not only in their own right, but also in that they constitute the principal parts in many other FPDEs. The main contribution of this work is threefold:
First, we introduce a new family of functional spaces defined by using fractional derivatives, and prove that these spaces are equivalent to usual Sobolev spaces in the sense that their norms are equivalent. Based on these spaces the variational formulation of the initial boundary value problems of FDEs are developed, and the existence and uniqueness of the weak solution are established by using classical theory for elliptic problems. The obtained results indicate that in the case of Riemann-Liouville definition, the equivalence between FDEs and weak formulation does not require any initial conditions. This contrasts with the case of Caputo definition, in which the initial condition has to be integrated into the weak formulation in order to establish the equivalence.
Second, based on the proposed weak formulation, we investigate the numerical solutions of the time fractional diffusion equation (TFDE). Essentially, the TFDE differs from the standard diffusion equation in the time derivative term. In TFDE, the first-order time derivative is replaced by a fractional derivative, making the problem global in time. We propose a spectral method in both temporal and spatial discretizations for this equation. The convergence of the method is proven by providing a priori error estimate. Numerical tests are carried out to confirm the theoretical results. Thanks to the spectral accuracy in both space and time of the proposed method, the storage requirement due to the ``global time dependence" can be considerably relaxed, and therefore calculation of the long-time solution becomes possible.
Third, we consider the fractional Nernst-Planck equation, which describes the anomalous diffusion in the movement of the ions in neuronal system. A method combining finite differences in time and spectral element methods in space is proposed to numerically solve the underlying problem. The detailed construction and implementation of the method are presented. Our numerical experiences show that the convergence of the proposed method is exponential in space and (2-alpha)-order (0< alpha < 1) in time. Finally, a practical problem with realistic physical parameters is simulated to demonstrate the potential applicability of the method.
Seminar: Test Theory and Rule Space Model
Speaker: Dr. Li, Feng, United International College
Time: 3:00-4:00 pm, Wed, Nov 11, 2009
Absract:
Classical Test Theory(CTT), Item Response Theory(IRT) and Cognition Diagnosis Model(CDM) are three stages in history of Test Theory. We discussed reliability, validity, difficulty and other critical ideas in Test theory and their operational definition in CTT and IRT. As one kind of important CDM, RSM can describe examinees' knowledge state and attribute master probability. In general, attribute means necessary knowledge and skill for some questions. According to results of the RSM, teachers and parents would help students in a effective way.
Seminar: The Pricing of Options and Futures
Speaker: Mr Fu, Song Feng, United International College
Time: 3:00-4:00 pm, Wed, Nov 4, 2009
Absract:
Firstly, introduce the notation of options and futures. Secondly, discuss two basic pricing methods for options: Black-Scholes model and binomial pricing model. Finally, introduce some new types of exotic options and the pricing formula is given. Then we are focused on barrier options and Asian options for application.
Seminar: Modified Tangential Frequency Filtering Decomposition and its Fourier Analysis
Speaker: Dr. Niu, Qiang, United International College
Time: 3:00-4:00 pm, Wed, Oct 28, 2009
Absract:
A modified tangential frequency filtering decomposition (MTFFD) preconditioner is proposed. The optimal order of the modification and the optimal relaxation parameter is determined by Fourier analysis. With the choice of optimal order of modification, the Fourier results show that the condition number of the preconditioned matrix is O(h^{-2/3}), and the spectrum distribution of the preconditioned matrix can be predicted by the Fourier results. The performance of MTFFD preconditioner is compared with tangential frequency filtering (TFFD) preconditioner on a variety of large sparse matrices arising from the discretization of PDEs with discontinuous coefficients. The numerical results show that the MTFFD preconditioner is much more efficient than the TFFD preconditioner.
Seminar: Estimate for Finite Time Ruin Probability with Insurance and Financial Risks
Speaker: Mr Zhou, Min, United International College
Time: 3:00-4:00 pm, Wed, Oct 21, 2009
Absract:
It is well known that the ruin theory is the key component of the risk probability and much attention has been paid to issues of ruin theory. One of the hot issues of this research is the asymptotic estimation of ruin probability of an insurer. Recently, a lot of papers have been published on the asymptotics estimates of ruin probability of an insurer who is exposed to a stochastic economic environment. This environment has two kinds of risk, which were called by Norberg(1999) as insurance risk and financial risk, respectively. Tang and Tsitsiashvili(2004) obtained several precise asymptotics estimates for the finite time ruin probability, in which insurance risk belongs to the intersection of the long-tailed distribution class and dominated-variation distribution class. Later, Chen and Xie(2005) used a simpler method to prove the same result under a weaker assumption on the financial risk. Under my tutor Professor Wang Yuebaos guidance and help, we give estimates for the finite time ruin probability with insurance and financial risks for two cases.
Seminar: COMPREHENSIVE FINANCIAL EVALUATION SYSTEM OF LISTED COMPANIES BASED ON DATA-DRIVEN
Speaker: Miss Lenon Li, Jian Xia, United International College
Time: 3:00-4:00 pm, Wed, Oct 14, 2009
Absract:
This paper is mainly researched at the performance of listed companies and giving a new method to this problem. Firstly, a comprehensive evaluation system of listed companies has been built in four aspects. The company scores have been calculated with Analysis Hierarchy Process method. Secondly, the weighting of financial indicators has been computed by using the "entropy" concept. On this basis, 670 listed companies were analyzed, then, we can get their rankings and scores.
Schedule of seminars for this semester
Date: Speaker:
14 October 2009 Li Jian Xia
21 October 2009 Zhou Min
28 October 2009 Niu Qiang
04 November 2009 Fu Songfeng
11 November 2009 Dr. Feng Li
18 November 2009 Dr. Xianjuan Li
25 November 2009 Dr. Lv Guangqing
10 ISCI members joined the 10th Chinese Conference on Uniform Design and Applied Statistics
10 members in ISCI attended the 10th Chinese Conference on Uniform Design and Applied Statistics held during July 13 to July 17, 2009 in Xi’an. The conference was organized by Xi’an University of Finance and Economics, the Uniform Design Association of China and UIC. In this conference, Prof. Fang gave an invited lecture “On Number-Theoretic Method in Statistics Simulation”. In addition, ISCI members gave six talks in this conference as follows:
a. "Comparison between different types of experimental designs" (given by Yongdao Zhou);
b. "E(f_NOD)" 最优混水平超饱和设计构造 (given by Haiyin Zhou);
c. "均匀实验设计的引用综述" (given by Guoqiu Zhang);
d. "分布式系统中H_∞滤波的最优融合" (given by Wenxuan Wang);
e. "Comparison of different distant matrices in clustering for distinguishing TCM fingerprints" (given by Xiaoling Peng);
f. "Comparison between trading style of institution and individual investor: evidence from china" (given by PingHe).
Seminar: 3He from Moon: A Safe and Clean Energy Source for Future Generations and the goal of the Chinese Chang'e Project
Speaker: Dr Tsang Kang-Too Ken , United International College
Time: 2:00-3:00 pm, Wed, Apr. 22, 2009
Absract:
Sometime around 2050, due to shortage of fossil fuels and their greenhouse effects, human society has to make the transition to other energy sources. Nuclear fission energy can provide a temporary solution but short of a permanent one because of the limited supply of fissible uranium and the proliferation problem that comes with it. Other forms of alternative energy like solar, wind or biofuels are either too expensive or diffuse in power density that they can only play a supplementary role. The only viable energy source that can assume a dominant role is nuclear fusion, if current international R&D effort in fusion technology is carried out as planned and achieving its goal successfully.
Of all fusion fuels that are technologically feasible in the near future, helium-3 is regarded as ideal judging from its high power conversion efficiency, low radioactivity, and its benign proliferation impact. However, terrestrial supply of helium-3 is extremely limited, because it rarely occurs in nature and is produced only as the decayed product of tritium. By one estimate, in year 2000 the total reserve of helium-3 on earth is just about 500 kg.
Significant amount of lunar helium-3 is well documented in literature. Helium-3 concentration from Apollo and USSR lunar soil samples are verified by scientific studies. Existing analyses published indicate that there are at least 1,000,000 tons of helium-3 imbedded on the lunar surface. This amount of lunar helium-3 is significant because it is estimated that fusion energy released from 30 tons of helium-3 is enough to satisfy the electric power consumed by USA for a year. There is roughly 10 times more energy contained in the lunar helium-3 than in all the economically recoverable fossil fuels on the Earth. Even at a cost of 1 billion US dollar per ton of helium-3, in terms of energy it is equivalent to oil at US$7 per barrel.
In this talk I will give an overall view of the issues involved, both scientific and non-scientific aspects of them. In particular, I will discuss how the Chinese Chang'e Project will contribute to the exploration of this important energy resource form the Moon.
Seminar: PROBABILITIES ESTIMATES FOR BROWNIAN PROCESS WITH POLYNOMIAL DRIFT AND TIME ESTIMATES FOR LEVY PROCESS ON P-ADICS
Speaker: Dr. Jiao Li, United International College
Time: 2:00-3:00 pm, Wed, April 1, 2009
Seminar:In search of the underlying structure in data: Facet Theory and multidimensional scaling as an alternative to factor analysis
Speaker: Arie Cohen, United International College
Time:2:30-4:00 pm, Wed, Nov 26, 2008
Absract:
Scientists in all disciplines attempt to identify general principles that govern various sets of phenomena or variables. In social sciences various methods have been offered in order to achieve this task. These attempts relate to the search for latent variables, factors, principal components or underlying structures. One comprehensive response to this issue is Facet Theory.
Facet Theory is a systematic approach for coordinating theory and research. This goal is achieved by employing mapping sentence - a mechanism for constituting a definitional framework for all relevant attributes of a phenomenon in order to construct empirical structural hypotheses – and testing its validity through multidimensional scaling.
In the lecture I will demonstrate few examples of mapping sentences and ways to validate them through the statistical procedure of multidimensional scaling. Next I will contrast multidimensional scaling as an alternative to factor analysis, focusing on conceptual and psychometric characteristics of these approaches and highlighting their differences in the context of several psychometric measures. Finally, I will demonstrate the use of facet theory in the process of developing a new scale for depression.
Seminar: Grouped Dirichlet Distribution: A New Tool for Incomplete Categorical Data Analysis
Speaker: Guo-Liang Tian, Division of Biostatistics, University of Maryland Greenebaum Cancer Center
Time: 2:30-3:30 pm, Friday, June 6, 2008
Absract: Motivated by the likelihood functions of several incomplete categorical data, this article introduces a new family of distributions, grouped Dirichlet distributions (GDD), which includes the classical Dirichlet distribution (DD) as a special case. First, we develop distribution theory for the GDD in its own right. Second, we use this expanded family as a new tool for statistical analysis of incomplete categorical data. Starting with a GDD with two partitions, we derive its stochastic representation that provides a simple procedure for simulation. Other properties such as mixed moments, mode, marginal and conditional distributions are also derived. The general GDD with more than two partitions is considered in a parallel manner. Three data sets from a case-control study, a leprosy survey, and a neurological study are used to illustrate how the GDD can be used as a new tool for analyzing incomplete categorical data. Our approach based on GDD has at least two advantages over the commonly used approach based on the DD in both frequentist and conjugate Bayesian inference: (a) In some cases, both the maximum likelihood and Bayes estimates have closed-form expressions in the new approach, but not so when they are based on the commonly-used approach; and (b) even if a closed-form solution is not available, the EM and data augmentation algorithms in the new approach converge much faster than in the commonly-used approach.
About Speaker:
Dr. Guo-Liang Tian is a senior biostatistician and faculty member in the division of Biostatistics, University of Maryland Greenebaum Cancer Center. His research interest covers areas of Incomplete categorical data analysis, Constrained parameter models and Variables selection, Sample surveys with sensitive questions, Non-iterative Monte Carlo methods, IBF Sampler and MCMC, Bayesian Analysis in Medicine, EM and MM algorithms, Reliability, prediction inferences and finance statistics and so on.
Seminar: Hierarchical Segmentation of Multimodality MRI Human Brain Tumors - Towards a Comprehensive Characterization of Tumors via Structural and Diffusion Tensor Magnetic Resonance Imaging
Speaker: Dr. Cai Hong-Min
Time: 2008. 05.21, 16:00-17:00
Abstract: The talk aims at creating a multi-modal profile of tissue components that will not only help in delineating tumor and edema from healthy tissue, distinguishing between enhancing and non-enhancing tumors, but also produce a probabilistic characterization of tissue around the tumor to determine abnormal regions that may have a tendency to convert to tumor in the future. The multimodality profile is generated by a combination of five structural MR images, FLAIR, T1, Gadolium enhanced T1 (GAD), DWI and B0, and two scalar maps, including Fractional Anisotropy(FA) and Apparant Diffusion Coefficient (ADC) computed from diffusion tensor images (DTI), creating a seven-dimensional intensity feature vector for each voxel. The tumor-grade- specific ground truth identified by doctors, are trained through a newly proposed hierarchical Support Vector Machine (SVM) based on spatial and texture features, the system achieves near-perfect characterization of tumor components (enhancing and non-enhancing), edema and healthy tissue with a 90  95 % classification rate together with almost 0 false positive rate. In addition to this hard tissue segmentation, the framework also provides probability profile for tissue, indicating unhealthy regions that may have a tendency to convert to tumorous tissue, hence providing a better characterization of the resection margin. The classifiers, trained on a dataset of 22 patients, can be applied to a new dataset with a success rate of 80% for classification, as has been tested using a leave-one-out paradigm on these 22 datasets. The multimodality processing pipeline that we have designed is general and is applicable to any study that has multi-modal data acquisition
About Speaker:
Dr. CAI Hongmin received his B.S and M.S degrees both from Harbin Institute of Technology, China. He obtained his PhD degree in applied mathematics from the University of Hong Kong in 2007. His research interest includes biomedical image analysis and biometric recognition.
Seminar: Two-level Multiple Discrete  Continuous Model
Speaker: Dr. Jiu-Kun Li
Time: 3:00-4:00pm,Wednesday, May 14th, 2008
Absract: This paper developed a two-level multiple discrete-continuous model. Details of the model and its estimation method are explained and illustrated by an example. The model may be used to investigate activity choices and time allocations in activity-travel behavior modeling; product and service choices and money allocation in marketing.
About Speaker:
Dr. Li Jiu Kun is currently a visiting scholar in UIC. She obtained her PhD degree from The University of Hong Kong. Her research interests are statistics and its applications, optimization, logistics and supply chain management, and transportation modeling. 
Seminar: Camera Calibration with Spheres: Linear Approaches
Speaker: Dr. Amy Zhang
Time: 3:00-4:00pm,Wednesday, Apr. 16th, 2008
Abstract: This paper addresses the problem of camera calibration from spheres. By studying the relationship between the dual images of spheres and that of the absolute conic, a linear solution has been derived from a recently proposed non-linear semi-definite approach. However, experiments show that this approach is quite sensitive to noise. In order to overcome this problem, a second approach has been proposed, where the orthogonal calibration relationship is obtained by regarding any two spheres as a surface of revolution. This allows a camera to be fully calibrated from an image of three spheres. Besides, a conic homography is derived from the imaged spheres, and from its eigenvectors the orthogonal invariants can be computed directly. Experiments on synthetic and real data show the practicality of such an approach.
Seminar:  Modeling of lunar Helium-3 distribution with remote sensing data from ChangˇE-1
Speaker: Prof. Tsang Kang Too
Time: 2008.04.02, 16:00-17:00
Abstract: Sometime around 2050, due to shortage of fossil fuels and their green house effects, human society has to make the transition to other energy sources. Nuclear energy can provide a temporary solution but short of a permanent one because of the limited supply of fissible uranium and the proliferation problem that comes with it. Other forms of alternative energy like solar, wind or biofuels are either too expensive or diffuse in power density that they can only play a supplementary role. The only viable energy source that can assume a dominant role is nuclear fusion, if current international R&D effort in fusion technology is carried out as planned and achieving its goal successfully.
HISTORY