We study nonlinear advection-diffusion problems of the form u_t - u_{xx} + /beta u_x = f(u) with free boundaries x = h(t) and x = g(t). Such a problem may be used to describe the spreading of a biological or chemical species with the free boundaries representing the expanding fronts. The term /beta u_x represents an effect of advective environments. For logistic nonlinearity, it has been shown by Gu, Lin and Lou that the asymptotic spreading speeds of the two fronts h(t) and g(t) are different due to the advection term. In this paper, when the nonlinear function is a monostable, bistable or of combustion type, we give a much sharper estimate for the different spreading speeds of the fronts, and describe how the solution approaches a semi-wave when spreading happens. We develop new approaches and extend the result of Du, Matsuzawa and Zhou to the problem with an advection term.

Auslander-Reiten theory for finite dimensional algebras is well-established. There is also Auslander-Reiten theory for orders, but it seems no references which define AR-quivers for orders in non semi-simple algebras and study their shapes. I will explain how to construct almost split sequences for self-injective orders and introduce Heller lattices. Then, we show, in a simple example, the shape of components of stable AR-quiver that contain Heller lattices.

Short-selling is important in asset–pricing and was legalized in China a few years ago. This topic would primarily suit a finance audience (post-graduates, staff and under-graduates) but might be of interest to staff and students from other disciplines too.

本报告中我们围绕第一性原理电子结构计算中一类非线性特征值问题展开分析，包括数值分析和数值模拟两方面。数值分析方面，我们研究了有限维逼近的收敛性，先验误差估计，后验误差估计，并且对这类非线性特征值问题设计了自适应有限元算法，分析了自适应有限元算法的收敛率和复杂度。数值模拟方面，我们围绕第一原理大型并行计算软件包ABINIT展开，实现了软件包ABINIT与交换关联泛函库LIBXC的接口，并利用ABINIT比较了不同交换关联泛函对不同物理性质的影响，为其他科研工作者选择合适的交换关联泛函提供了依据

同济大学特邀讲座报告主题金融研究方向的过去、现在和未来Research in Finance : Past, Present and Future报告嘉宾：金融传奇教父John Hull教授加拿大多伦多大学罗特曼管理学院报告时间：3月12日下午2：00 - 3：30报告地点：同济大厦A楼301报告厅报告人简介：John Hull教授是加拿大多伦多大学罗特曼管理学院的金融学资深教授。自上世纪八十年代以来他一直关注于金融风险和衍生品研究，是全球金融学定价理论和风险计量领域中的...

In this talk I will first give an overview of the family of semiparametric transformation models and methods for handling such models. I will then present some biased sampling schemes and introduce a general approach to the transformation models under biased sample schemes. Finally I will discuss asymptotic theory and numerical results on some data sets.

Stochastic models are becoming a more and more important tool to describe complex natural systems. In finance and economics, stochastic differential equation is a fundamental method to study the uncertainty in financial systems. A major challenge in stochastic modeling is how to estimate unknown model parameters based on the financial market data. In this talk we will discuss two types of major methods for estimating model parameters: the optimization method and Bayesian inference methods. These methods will be applied to infer parameters in financial models.

Theory of neutral (i.e., optimal) and indifference pricing, hedging and optimal trading of portfolios of financial contracts, for completely general diffusive Markovian continuous-time financial models is nowadays available, due to the works of the speaker. Furthermore, this methodology is fully implemented using symbolic calculations on Mathematica computer platform. As a consequence, financial engineering solutions of unprecedented complexity can nowadays be achieved in both, complete and incomplete markets. This talk will showcase the above claims on some examples in commodities and commodity derivatives. Many other application areas, such as equity hedge fund management, foreign exchange and foreign exchange derivatives are also available