No description available.

In some survival analysis of medical studies, there are often long term survivors who can be considered as permanently cured. The goals in these studies are to estimate the cure probability of the whole population and the hazard rate of the noncured subpopulation. The existing methods for cure rate models have been limited to parametric and semiparametric models. More specifically, the hazard function part is estimated by parametric or semiparametric model where the effect of covariate takes a parametric form. And the cure rate part is often estimated by a parametric logistic regression model. We introduce a non-parametric model employing smoothing splines. It provides non-parametric smooth estimates for both hazard function and cure rate. By introducing a latent cure status variable, we implement the method using a smooth EM algorithm. Louis' formula for covariance estimation in an EM algorithm is generalized to yield point-wise confidence intervals for both functions. A simple model selection procedure based on the Kullback-Leibler geometry is derived for the proposed cure rate model. Numerical studies demonstrate excellent performance of the proposed method in estimation, inference and model selection. The application of the method is illustrated by the analysis of a melanoma study. / Ph. D.

Ablative materials can sustain very high temperatures in which surface thermochemical processes are significant enough to cause surface recession. Existence of moving boundary over a wide range of temperatures, temperature-dependent thermophysical properties of ablators, and no prior knowledge about the location of the moving surface augment the difficulty for predicting the exposed heat flux at the receding surface of ablators. In this paper, the conjugate gradient method is proposed to estimate the unknown surface recession and time-varying net surface heat flux for these kinds of problems. The first order Tikhonov regularization is employed to stabilize the inverse solution. Considering the complicated phenomena that are taking place, it is shown via simulated experiment that unknown quantities can be obtained with reasonable accuracy using this method despite existing noises in the measurement data.

No description available.

A simulation study consists of several steps such as data collection, coding and model verification, model validation, experimental design, output data analysis, and implementation. Our research concentrates on output data analysis. In this field, many researchers have studied how to construct confidence intervals for the mean u of a stationary stochastic process. However, the estimation of the value of a nonlinear function f(u) has not received a lot of attention in the simulation literature. Towards this goal, a batch-means-based methodology was proposed by Munoz and Glynn (1997). Their approach did not consider consistent estimators for the variance of the point estimator for f(u). This thesis, however, will consider consistent variance estimation techniques to construct confidence intervals for f(u). Specifically, we propose methods based on the combination of the delta method and nonoverlapping batch means (NBM), standardized time series (STS), or a combination of both. Our approaches are tested on moving average, autoregressive, and M/M/1 queueing processes. The results show that the resulting confidence intervals (CIs) perform often better than the CIs based on the method of Munoz and Glynn in terms of coverage, the mean of their CI half-width, and the variance of their CI half-width.

The statistical properties of cross-validation bandwidths can be improved by choosing an appropriate kernel, which is different from the kernels traditionally used for cross- validation purposes. In the light of this idea, we developed two new methods of bandwidth selection termed: Indirect cross-validation and Robust one-sided cross- validation. The kernels used in the Indirect cross-validation method yield an improvement in the relative bandwidth rate to n^1=4, which is substantially better than the n^1=10 rate of the least squares cross-validation method. The robust kernels used in the Robust one-sided cross-validation method eliminate the bandwidth bias for the case of regression functions with discontinuous derivatives.

As the trading volume in TURKDEX, the first and only options and futures exchange in Turkey, increases, it becomes more important to have an understanding of the effect of stock index futures trading on the underlying spot market volatility. In this respect, this thesis analyzes the effect of ISE-National 30 index futures contract trading on the underlying stocks&rsquo / volatility. In this thesis, spot portfolio volatility is decomposed into two components and this decomposition is applied to a single-factor return-generating model to focus on the relationships among the volatility components rather than on the components in isolation. In order to measure the average volatility and the cross-sectional dispersion of the component securities and the portfolio volatility for each day in the sample period, a simple filtering procedure to recover a series of realized volatilities from a discrete time realization of a continuous time diffusion process is used. Results reveal that inception of futures trading has no significant effect on the volatility of the underlying ISE National 30 index stock market.

Reducing power consumption is among the top concerns in Wireless Sensor Networks, as the lifetime of a Wireless Sensor Network depends on its power consumption. Directional antennas help achieve this goal contrary to the commonly used omnidirectional antennas that radiate electromagnetic power equally in all directions, by concentrating the radiated electromagnetic power only in particular directions. This enables increased communication range at no additional energy cost and reduces contention on the wireless medium. The SPIDA (SICS Parasitic Interference Directional Antenna) prototype is one of the few real-world prototypes of electronically switchable directional antennas for Wireless Sensor Networks. However, building several prototypes of SPIDA and conducting real-world experiments using them may be expensive and impractical. Modeling SPIDA based on real-world experiments avoids the expenses incurred by enabling simulation of large networks equipped with SPIDA. Such a model would then allow researchers to develop new algorithms and protocols that take advantage of the provided directional communication on existing Wireless Sensor Network simulators. In this thesis, a model of SPIDA for Wireless Sensor Networks is built based on thoroughly designed real-world experiments. The thesis builds a probabilistic model that accounts for variations in measurements, imperfections in the prototype construction, and fluctuations in experimental settings that affect the values of the measured metrics. The model can be integrated into existing Wireless Sensor Network simulators to foster the research of new algorithms and protocols that take advantage of directional communication. The model returns the values of signal strength and packet reception rate from a node equipped with SPIDA at a certain point in space given the two-dimensional distance coordinates of the point and the configuration of SPIDA as inputs. / Phone:+46765816263 Additional email: burkaja@yahoo.com

In this thesis, we study the estimation of the nonlinear conditionalrisk-neutral density function (RND) in discrete time. Specifically, weevaluate the extent to which the estimated nonlinear conditional RNDvaluable insights to answer relevant economic questions regarding to optionpricing, the measurement of invertors' preferences and portfolio choice.We make use of large dataset of options contracts written on the S&P 500index from 1996 to 2011, to estimate the parameters of the conditional RNDfunctions by minimizing the squared option pricing errors delivered by thenonlinear models studied in the thesis.In the first essay, we show that a semi-nonparametric option pricing modelwith GARCH variance outperforms several benchmarks models in-sample andout-of-sample. In the second essay, we show that a simple two-state regimeswitching model in volatility is not able to fully account for the pricingkernel and the risk aversion puzzle; however, it provides a reasonablecharacterisation of the time-series properties of the estimated riskaversion.In the third essay, we evaluate linear stochastic discount factormodels using an out-of-sample financial metric. We find that multifactormodels outperform the CAPM when this metric is used, and that modelsproducing the best fit in-sample are also those exhibiting the bestperformance out-of-sample.

This study intends to build up new techniques for how to obtain completely data-driven choices of the smoothing parameter in functional estimation, within the confines of minimal assumptions. The focus of the study will be within the framework of the estimation of the distribution function, the density function and their multivariable extensions along with some of their functionals such as the location and the integrated squared derivatives.