Abstract
SIMEX is a general-purpose technique for measurement error correction. There is a substantial literature on the application and theory of SIMEX for purely parametric problems, as well as for purely nonparametric regression problems, but there is neither application nor theory for semiparametric problems. Motivated by an example involving radiation dosimetry, we develop the basic theory for SIMEX in semi- parametric problems using kernel-based estimation methods. This includes situations that the mismeasured variable is modeled purely parametrically, purely nonparametrically, or that the mismeasured variable has components that are modeled both parametrically and nonparametrically. Using our asymptotic expansions, easily computed standard error formulae are derived, as are the bias properties of the nonparametric esti- mator. The standard error method represents a new method in general for semiparametric problems, and we show in our example that it improves dramatically on the first order methods.

Abstract
Several goodness-of-fit tests are proposed for testing the goodness-of-fit of discrete multivariate data. We propose a unified analysis of goodness-of-fit using a generalized test statistic(s). Depending on the choice of parameters, the generalized test statistics results into some of the known statistics such as Chisquare, and likelihood ratio generalized test, it has been observed that the proposed test statistic is highly robust against extreme values and does not assume the distribution of parent population. The asymptotic distribution of the proposed test statistic and the p-value function are discussed. The power comparisons have been made, and it is found that, the proposed statistic dominates over the Cressie and Read (1984) test statistic however the comparisons are made with respect to change in the location only. The application of proposed method has been attempted by using a real-life data, of Cranor and Christensen.

Abstract
Bayesian bootstrapping (BBS, in short) in a broad sense, including nonparametric bootstrapping as a special case, is a versatile method for estimating the sampling distribution of an estimator T(Fn) of a population parameter T(F). Here, Fn is the empirical distributions based on observations of size n, F is the (unknown) population and T is a (known) functional with gentle characters typically written as a C-infinity function of moments defined in a neighborhood of the population moments. Assume that there exists an (unknown) functional TF such that T(F) = TF (U) for a (well known) distribution U, e.g., the uniform or normal distributions, then approximation to U by BBS from Un, i.e., a random sample of size n from U, should be our main concern, since T(Fn) and TF (Un) have the same sampling distribution. Algebraic and numerical works from this viewpoint have revealed that BBS with the Dirichlet prior Dir[n; c; :::; c] for c around 1/2 gives robust approximation to the target distribution. Numerical comparisons are also made with several resampling methods.

Abstract
Based on reproducing-kernel Hilbert space theory, a large class of feature spaces can be characterized through their kernel functions, which have been widely applied in statistics and engineering. By further parameterizing the kernel structure, many current nonparametric regularization methods can be extended to achieve both model atting and feature selection objectives. In this paper, we focus on the nonparametric regularization problems associated with a kernel collection, and introduce a fast adaptive kernel selection algorithm rooted in parametric linear programming.

Abstract
A research is conducted in order to determine the risk of contracting cancer of the children. The sample is defined as the children who have been treated in Gazi University Hospital, Department of Children Health and Diseases. The factors that can be effective as a risk of contracting cancer among the children have been listed and regarding that, a questionnaire is formed. The survey is conducted to 616 children, 233 girls and 383 boys. The survey lasted 1.5 years. The dependent variable for binary logistic model was defined as contracting cancer or not. As a result of logistic regression, the significant variables in the model are mothers age, mothers education, breast feeding duration, and if there is a cancer occurrence in the family or not. The logistic regression model for the risk of contracting cancer among the children is constructed, and interpreted. Also the cell probabilities and prediction values are given in a resulting table.