Friday, April 24, 2009
On the Fundamental properties of Fractional Brownian Motion process and applications
In recent years fractional Brownian motion (fBM) process has been suggested to replace the classical Brownian motion process as driving process in the modeling of many real world phenomena, including stock price modeling, polymer molecule, etc. This talk focuses on some fundamental properties of such process and its applications, then the Ito formula.
Dynamic Insurance Risk Models
We first introduce the classical renewal model and its special case, the Cramer-Lundberg model. Moreover, the extension of this model is also outlined. The properties of the Cramer-Lundberg model is discussed, namely the expected value of the surplus at time \(t\), its ruin probability and determination of the initial capital. The renewal model is then re-formulated into a dynamic system. The role and scope of the dynamic system are explored.
Friday, April 17, 2009
Stochastic modeling of Network Centric Epidemiological Processes
We discuss the formulation of a dynamic model for the spread of infectious diseases on a heterogeneous population via network approach. Highlighting important network properties namely degree distribution, clustering coefficient and average shortest path length. These properties characterize the method and choice of the network models; for example, a scale-free network model for an infectious disease transmission that confers permanent immunity upon recovery in an infinite closed population. We propose to apply statistical techniques to analyze the model and discuss possible important results.
The Stochastic Model for Network Externality in Market Structure
There are many factors to explain the failure of market such as adverse selection, moral hazard, externality, etc. The network externality is an interesting factor of market failure. How does the network externality under random environmental perturbations affect the demand function, market equilibrium, and firm’s behavior? In this presentation, a part of my dissertation, I will develop a stochastic demand function for network externality in market.
Friday, April 10, 2009
The Frontiers of Theory and Application of Stocastic Dynamic-Hybrid Systems
Founder, IFNA and GVP-V. Lakshmikantham Institute of Advanced Studies, India
Friday, April 3, 2009
Red Tide and it’s Biological Effects
The present study considers the random phenomenon that is Red Tide as found around the State of Florida. Among the many organism that make up Red Tide, Karenia Brevis is the organism commonly associated with an outbreak, the probability distribution which best describes the behavior of Karenia Brevis is the Weibull probability distribution. There are regional differences as well as regional relationships including delay effects. Recursion rates indicate a logistic growth model; however, additional information is needed before research can determine the effects of runoff on Red Tide blooms.
Friday, March 27, 2009
- Spring Picnic
Friday, March 13, 2009
No seminar this week.
Friday, February 27, 2009
Some Supervised Learning Methods for Classification
Nabin Manandhar Shrestha
In this talk I will discuss about some popular supervised learning methods for the classification and its limitation when applying to microarray data. Furthermore, I will show that how the genes selected by the BF method improves the classsification performance of the classifiers.
Friday, February 20, 2009
This week's seminar has been replaced by an Interdisciplinary Workshop.
Friday, February 13, 2009
Parametric and Nonparametric Analysis and Modeling of Breast Cancer, Survival Analysis of Breast Cancer, Nonparametric Survival Analysis for Cancer Data, Generalized Modeling in Survival Analysis — Breast Cancer and Statistical Modeling of Water Quality
Chunling Cong, Dimitrios Vovoras, Yong Xu and Keith Hackett
Various topics that will be presented to our guest Dr. James Kepner, the president of the America Cancer Society during the Interdisciplinary Workshop, Friday February 20, 2009.
Friday, February 6, 2009
Flexible covariate effects in the proportional hazards model
The proportional hazards model, often used to analyze the results of clinical trials when the outcomes are right censored, is used to outline a class of semiparametric models. In generalized additive models the effect of the covariate is additive and is represented by a smooth function. They allow nonlinear modeling of the prognostic factors and leave the data suggest the form of their effect. The methods are illustrated in an analysis from a breast cancer clinical trial.
Friday, January 30, 2009
Nonlinear Stochastic Modeling and Statistical Analysis
We will first review linear stochastic modeling briefly. Then we will show our algorithms to constract nonlinear stochastic models. We will also present the simulation results using different data partitions. Then we will statistically analyze the jumps, with respect to the length between jumps and the magnitude of jumps.