The Objective of the Remote Measurements Calibration (RMC) method is to minimize systematic errors through an appropriate scaling procedure. A new method for RMC has been developed. This method solves the problems of observability, multiplicity of solutions, and ambiguity of reference points associated with the method proposed by Adibi et. al. [6-9]. The new algorithm uses the simulated annealing technique together with the matroid method to identify and minimize the number of RTUs (Remote Terminal Units) required to observe the system. After field calibration, these RTUs provide measurements that are used to estimate the whole state of the system. These estimates are then returned as a reference for remotely calibrating the remaining RTUs. The calibration coefficients are estimated by means of highly robust estimator, namely the Least Median of Squares (LMS) estimator. The calibration method is applicable to large systems by means of network tearing and dynamic programming. The number of field calibrations can be decreased further whenever multiple voltage measurements at the same buses are available. The procedure requires that the measurement biases are estimated from recorded metered values when buses, or lines, or transformers are disconnected. It also requires the application of a robust comparative voltage calibration method. To this end, a modified Friedman test has been developed and its robustness characteristics investigated. / Ph. D.

The purpose of this thesis is to compare rank-based tests of bivariate stochastic order. Given two bivariate distributions $F$ and $G$, the general problem we are dealing with is to test $H_0: F=G$ against $H_1:F<G$, where $F$ and $G$ are independent continuous distributions on $\Re ^2$. (``$F<G$" means that $F(x)\leq G(x)~\forall x\in \Re^2$, and $\exists x\in \Re^2$ such that $F(x)< G(x)$.). In particular, we will analyze three analogues of the one-dimensional Mann-Whitney-Wilcoxon test in two dimensions. Two of the test statistics are new; we call them the Kendall and Spearman statistics. We will then show the asymptotic distributions and carry out empirical comparisons of the Kendall, Spearman and the third two-dimensional Mann-Whitney-Wilcoxon statistics.

The purpose of this thesis is to compare rank-based tests of bivariate stochastic order. Given two bivariate distributions $F$ and $G$, the general problem we are dealing with is to test $H_0: F=G$ against $H_1:F<G$, where $F$ and $G$ are independent continuous distributions on $\Re ^2$. (``$F<G$" means that $F(x)\leq G(x)~\forall x\in \Re^2$, and $\exists x\in \Re^2$ such that $F(x)< G(x)$.). In particular, we will analyze three analogues of the one-dimensional Mann-Whitney-Wilcoxon test in two dimensions. Two of the test statistics are new; we call them the Kendall and Spearman statistics. We will then show the asymptotic distributions and carry out empirical comparisons of the Kendall, Spearman and the third two-dimensional Mann-Whitney-Wilcoxon statistics.

The purpose of this thesis is to compare rank-based tests of bivariate stochastic order. Given two bivariate distributions $F$ and $G$, the general problem we are dealing with is to test $H_0: F=G$ against $H_1:F<G$, where $F$ and $G$ are independent continuous distributions on $\Re ^2$. (``$F<G$" means that $F(x)\leq G(x)~\forall x\in \Re^2$, and $\exists x\in \Re^2$ such that $F(x)< G(x)$.). In particular, we will analyze three analogues of the one-dimensional Mann-Whitney-Wilcoxon test in two dimensions. Two of the test statistics are new; we call them the Kendall and Spearman statistics. We will then show the asymptotic distributions and carry out empirical comparisons of the Kendall, Spearman and the third two-dimensional Mann-Whitney-Wilcoxon statistics.

The purpose of this thesis is to compare rank-based tests of bivariate stochastic order. Given two bivariate distributions $F$ and $G$, the general problem we are dealing with is to test $H_0: F=G$ against $H_1:F<G$, where $F$ and $G$ are independent continuous distributions on $\Re ^2$. (``$F<G$" means that $F(x)\leq G(x)~\forall x\in \Re^2$, and $\exists x\in \Re^2$ such that $F(x)< G(x)$.). In particular, we will analyze three analogues of the one-dimensional Mann-Whitney-Wilcoxon test in two dimensions. Two of the test statistics are new; we call them the Kendall and Spearman statistics. We will then show the asymptotic distributions and carry out empirical comparisons of the Kendall, Spearman and the third two-dimensional Mann-Whitney-Wilcoxon statistics.

In cluster analysis, a fundamental problem is to determine the best estimate of the number of clusters, which has a deterministic effect on the clustering results. However, a limitation in current applications is that no convincingly acceptable solution to the best-number-of-clusters problem is available due to high complexity of real data sets. In this dissertation, we tackle this problem of estimating the number of clusters, which is particularly oriented at processing very complicated data which may contain multiple types of cluster structure. Two new methods of choosing the number of clusters are proposed which have been shown empirically to be highly effective given clear and distinct cluster structure in a data set. In addition, we propose a sequential type of clustering approach, called multi-layer clustering, by combining these two methods. Multi-layer clustering not only functions as an efficient method of estimating the number of clusters, but also, by superimposing a sequential idea, improves the flexibility and effectiveness of any arbitrary existing one-layer clustering method. Empirical studies have shown that multi-layer clustering has higher efficiency than one layer clustering approaches, especially in detecting clusters in complicated data sets. The multi-layer clustering approach has been successfully implemented in clustering the WTCHP microarray data and the results can be interpreted very well based on known biological knowledge. Choosing an appropriate clustering method is another critical step in clustering. K-means clustering is one of the most popular clustering techniques used in practice. However, the k-means method tends to generate clusters containing a nearly equal number of objects, which is referred to as the ``equal-size'' problem. We propose a clustering method which competes with the k-means method. Our newly defined method is aimed at overcoming the so-called ``equal-size'' problem associated with the k-means method, while maintaining its advantage of computational simplicity. Advantages of the proposed method over k-means clustering have been demonstrated empirically using simulated data with low dimensionality. / Ph. D.

In this paper, I'll introduce the applications of Mathematica 7th and 8th edition in probability and statistics. The major contents are statistical data and charts, basis statistics, probability distributions, hypothesis testing, distribution fitting, regression analysis, cluster analysis and so on. Except introducing variously related commands, this paper will provides corresponding examples, so it can be regarded as a toolbook for people interested in the probabilities and statistical parts of this software.

Student's t Distribution is introduced with background including comments on its robustness properties. The ad hoc procedure of pretesting data for normality is discussed in the light of current advice, and previous work into its effectiveness reviewed. The approach to the problem is outlined: √b1 for test statistic, the Gram-Charlier distribution for population, approximations using the Johnson system.

In an attempt to develop a statistical model for fingerprint analysis, the Minutiae:Ridge-Length Ratio (MRLR) was used to examine differences between the different sexes, hands, fingers, ridge patterns, and racial groups. In regards to sex, statistically significant MRLR differences (α=0.05) were found between males and females when analyzing both individual prints (p<0.001) and entire ten-print cards (p=0.025). Further examination of the data revealed more specific differences within separate subcategories. The MRLR was significantly different (p<0.001) between males and females with both right and left hands. With respect to sex and individual finger differences, the thumb (p<0.001), index finger (p<0.001), and middle finger (p=0.015) were statistically significant. For ridge pattern, whorls (p<0.001) and ulnar loops (p<0.001) had significant differences between the sexes. Racially, males and females had statistically significant differences from one another within the Caucasian (p<0.001) and African American (p<0.001) racial groups. Further investigation of variables independent of sex highlighted other statistically significant MRLR relationships. Within the fingers, the thumb was found to be significantly different than the middle (p<0.001), ring (p<0.001), and little fingers (p<0.001); the index finger also differed from the little finger significantly (p=0.001). Comparison of level one detail demonstrated the whorl pattern was statistically different than the arch (p<0.001), radial loop (p=0.002), and ulnar loop patterns (p<0.001). No statistically significant difference was found between the right and left hands of the sample population (p=0.160). The racial subdivisions produced more complex relationships. Caucasians had statistically significant MRLR differences to African Americans (p=0.036), Hispanics (p=0.003), and Asians (p=0.046). African Americans had additional significant differences from Hispanics (p<0.001), Asians (p<0.001), and Native Americans (p=0.036). Finally, Native Americans and Hispanics shared a significant difference as well (p<0.036). However, due to the uncertainty of racial demographic data, the extrapolation of these findings to the general population may not be appropriate for forensic investigation purposes.

The purposes of the thesis were to review some of the existing methods for testing normality and to investigate the use of generated data combined with observed to test for normality. The approach to testing for normality is in contrast to the existing methods which are derived from observed data only. The test of normality proposed follows a characterization theorem by Bernstein (1941) and uses a test statistic D*, which is the average of the Hoeffding’s D-Statistic between linear combinations of the observed and generated data to test for normality. Overall, the proposed method showed considerable potential and achieved adequate power for many of the alternative distributions investigated. The simulation results revealed that the power of the test was comparable to some of the most commonly used methods of testing for normality. The test is performed with the use of a computer-based statistical package and in general takes a longer time to run than some of the existing methods of testing for normality.