There are two sources of information available in empirical research in finance: one corresponding to historical data and the other to prices currently observed in the markets. When proposing a model, it is desirable to use information from both sources. However in modern finance, where stochastic differential equations have been one of the main modeling tools, the common models are typically different for historical data and for current market data. The former are usually assumed to be time homogeneous, while the latter are typically time in-homogeneous. This practice can be explained by the fact that a time-homogeneous model is stationary and easier to estimate, while time-inhomogeneous model are required in order to replicate market data sufficiently well without creating arbitrage opportunities. In this thesis, we study methods of statistical inference, both parametric and non-parametric, for stochastic differential equations with time-dependent parameters. In the first part, we propose a new class of stochastic differential equation with time-dependent drift and diffusion terms, where some of the parameters change according to a hidden Markov process. We show that under some technical conditions this innovative way of modeling switching times renders the resulting model stationary. We also explore different approaches to estimate parameters in our proposed model. Our simulation studies demonstrate that the parameters of the model can be efficiently estimated by using a version of the filtering method proposed in the literature. We illustrate our model and the proposed estimation method by applying them to interest rate data, and we detect significant time variations in early 1980s, when targets of the monetary policy in the United States were changed. One of the known drawbacks of parametric models is the risk of model misspecification. In the second part of the thesis, we allow the drift to be time-dependent and nonparametric, and our objective is to estimate it using a single trajectory of the process. The main idea underlying this method is to approximate the time-dependent function with a sequence of polynomials. Since we can estimate efficiently only a finite number of parameters for any finite length of data, in our method we propose to relate the number of parameters to the length of the observed trajectory. This idea is similar to the method of sieves proposed by Grenander (Abstract Inference, 1981). The asymptotic analysis that we present is based on the assumption that the length of available data $T$ increases to infinity. We investigate two cases, one is a Brownian motion with time-dependent drift and the other corresponds to a class of mean-reverting stochastic differential equations with time-dependent mean-reversion level. In both cases we prove asymptotic consistency and normality of a modified maximum likelihood estimator of the projected time-dependent component. The main challenge in proving our results in the second case stems from two features of the problem: one is due to the fact that coefficients of projections change with $T$ and the other is related to the confounding effect between the mean-reversion speed and the level function. By applying our method to the same interest rate data we use in the first part, we find another evidence of time-variation in the drift term.

none

Tight gas reservoirs are those reservoirs where the matrix has a low permeability range (k < 0.1 md). The literature documents laboratory experiments under restressed conditions that show stress dependent rock properties are more significant in tighter rocks. For gas reservoirs, real gas properties are also sensitive to variations of pressure, and the correct description of gas flow must include pressure-dependent gas properties. Under these circumstances the resulting equation for real gas flow is a second order, non-linear, partial differential equation. Non-linearities include pressure-dependence of gas viscosity, gas compressibility, reservoir permeability and reservoir porosity. This paper investigates dynamic permeability change as a function of net overburden stress in tight gas reservoirs. The gas reservoir simulator used for this work included pressure-dependent reservoir permeability. Radial flow cases are analyzed using this simulator. During this study we found that from analysis of production data alone, it is impossible to determine the correct permeability value for tight gas reservoirs with pressure-dependent permeability. For the cases studied, the transient performance was similar for both constant permeability and pressure-dependent permeability. This similarity causes constant permeability and pressure-dependent permeability to be indistinguishable, based on analysis of transient performance data. It was found that the productivity index decreases when pressure-dependent permeability is more significant. Finally, this study verified that the method of Ibrahim et al.28 under estimates original gas in place (OGIP) for tight gas reservoirs with pressure-dependent permeability.

Modern digital designs require high performance and low cost. In this scenario, timing analysis is an essential step for each phase of the integrated circuit design cycle. To minimize the design turn-around time, the ability to correctly predict the timing behavior of the chip is extremely important. This has resulted in a demand for techniques to perform an accurate timing analysis. A number of existing timing analysis approaches are available. Most of these are pessimistic in nature due because of some inherent inaccuracies in the modeling of the timing behavior of logic gates. Although some techniques use accurate gate delay models, they have only been used to calculate the longest sensitizable delay or the shortest topological path delay for the circuit. In this work, a procedure to and the shortest destabilizing delay, as well as the longest sensitizable delay of a static CMOS circuit is developed. This procedure is also able to determine the exact circuit path as well as the input vector transition for which the shortest destabilizing (or longest sensitizable) delay can be achieved. Over a number of examples, on an average, the minimum destabilizing delay results in an improvement of 24% as compared to the minimum static timing analysis approach. The maximum sensitizable timing analysis results in an improvement of 7% over sensitizable timing analysis with pin-to-output delays. Therefore, the results show that the pessismism in timing analysis can be considerably decreased by using data dependent gate delays for maximum as well as minimum sensitizable timing analysis.

Fish nodavirus causes the death of several high economic fish. For studying the package and stability of dragon grouper necrosis virus (DGNNV), wild type virus-like particles (wt-VLPs) and £GN25-VLPs were employed. After experiments of disassembly and assembly, the result showed that DNA of 608 bp was able to be packaged by wt-VLPs. The sedimentation of VLPs was affected by different pH buffers during the process of purification: the VLPs at alkaline conditions behaved faster than those at the acidic. The stability of pure VLPs was pH-independent. The micrographs of the wt-VLPs in alkaline buffers showed that particles were rough and irregular in shapes, some of which were stain-permeable. However, the wt-VLPs in acidic buffers were morphologically indistinguishable from the untreated VLPs. The Western blotting for gradiently purified miture of VLPs and lysozyme revealed that lysozyme was co-precipitated with wt-VLPs.

According to the report of the Bureau of National Healthcare Insurance (NHI) in 1997, the total expenses on ventilator-dependent patients was about 7,100 million yuan in hospital, account for 3% of the cost of one year of health insurance of the whole people. To efficiently control their admission so as to decrease unsuitable utilization of mechanical ventilation, and to achieve the rational growth of medical expenditure, the NHI has developed the perspective payment system for the ventilator-dependent integrated delivery system (IDS) respiratory care program since July 1, 2000. Ventilator-dependent patients, difficult to wean, rely on the mechanical ventilation, using for at least 21 days in succession. The patients are dependent upon long-term mechanical respiratory care. Integrating the different level of respiratory care, IDS program is including ¡§ICU , intensive care unit¡¨, ¡¨RCC, respiratory care center¡¨, ¡¨ RCW , respiratory care ward¡¨ and ¡§home care¡¨ and pay in accordance with the level. The purpose of IDS program is to promote the quality of respiratory care and effectively to utilize the limited medical resources. The data for this research was retrieved from the 2002-2004 ¡§NHI database¡¨ that includes charge and discharge information for 115 hospitals. Of the 115 hospitals analyzed using data envelopment analysis (DEA) technique, to explore the whole efficiency and purely technological efficiency.

Thesis (Ph. D.)--University of Texas at Austin, 2000. / Vita. Includes bibliographical references (leaves 120-134). Available also in a digital version from Dissertation Abstracts.

Thesis (Ph. D.)--Rutgers University, 2009. / "Graduate Program in Nursing." Includes bibliographical references (p. 81-87).

Thesis (M.S.)--Ball State University, 2009. / Title from PDF t.p. (viewed on Nov. 30, 2009). Includes bibliographical references (p. [59]-63).

Thesis (M. Phil.)--Hong Kong University of Science and Technology, 2003. / Includes bibliographical references (leaves 127-148). Also available in electronic version. Access restricted to campus users.