Identifying landuse management practices is important for detecting landuse change and impacts on the surrounding landscape. The Ontario Ministry of Agriculture and Rural A airs has established a database product called the Agricultural Resource Inventory (AgRI), which is used for the storage and analysis of agricultural land management practices. This thesis explores the opportunity to populate the AgRI. A comparison of two supervised classi fications using optical satellite imagery with multiple single-date classifi cations and a subsequent multi-date, multi-sensor classi fication were used to gauge the best image timing for crop classi fication. In this study optical satellite images (Landsat-5 and SPOT-4/5) were inputted into a decision tree classifi er and Maximum Likelihood Classifi er (MLC) where the decision tree performed better than the MLC in overall and class accuracies. Classifi cation experienced complications from visual diff erences in vegetation. The multi-date classifi cation performed had an accuracy of 66.52%. The lack of imagery available at crop ripening stages reduced the accuracies greatly.

Markovo proceso tikimybinio mato absoliutaus tolydumo nesudėtingos sąlygos leidžia gauti atitinkamų statistinių eksperimentų tikėtinumo santykio pavidalą, kurio asimptotinės savybės susijusios su dviejų paprastų hipotezių asimptotiniais atskyrimo uždaviniais, kai yra taikomas maksimalaus tikėtinumo arba minimakso kriterijus. Tų paprastų hipotezių asimptotinis atskyrimas yra charakterizuojamas 1-os ir 2-os rūšies klaidos tikimybėmis, kurių asimptotinis elgesys priklausomai nuo optimalaus statistinio kriterijaus parinkimo užsirašo dvejomis formulėmis. Maksimalaus kriterijaus atveju tokia formulė buvo gauta bendriausiu atveju, tik nebuvo pritaikyta Markovo procesui su dideliu būsenų skaičiumi. Šiame darbe kaip tik parodyti šie taikymai. Taikant maksimalaus tikėtinumo kriterijų (Neimono-Pirsono) atitinkamas rezultatas buvo gautas tik tuo atveju, kai stebėjimai yra nepriklausomi ir vienodai pasiskirstę. Analogiškas rezultatas gautas bendriausiu atveju – gautos sąlygos, kada galioja atitinkama asimptotinė formulė. Kartu, pavyzdžiuose yra parodyti šios asimptotinės formulės taikymai, kai stebimas Markovo procesas su dideliu būsenų skaičiumi. / Absolute continuity simple conditions of probabilistic measure of Markov process allows you to get relevand statistical experiments likelihood ratio form, which asymptotic properties is associated with the asymptotic separation of the two simple hypotheses tasks, when is applied maximum likelihood (Neiman-Pirson) or minimax criterion. That asymptotic separation of the two simple hypothesis is characterized by type I and type II errors of probability, which asymptotic behavior depending on the optimal statistical criterion selection note down by two formulas. In maximum likelihood criterion case, formula was obtained on a very general case, not only been applied of the Markov process with a large number of states. These applications are shown at this work. Using maximum likelihood criterion (Neiman-Pirson) corresponding result was obtained only in that case, when observations are independent and identically distributed. Analogous result were obtained on a very general case – from conditions, when is valid the asymptotic formula. In examples of this work are shown that asymptotic formula applications, when is observed Markov process with a large number of states.

<p>The intention is to draw more specific connections between certain deconvolution methods and also to demonstrate the application of the statistical theory of estimation in the presence of measurement error. A parametric methodology for deconvolution when the underlying distribution is of the Pareto form is developed. Maximum likelihood estimation (MLE) of the parameters of the convolved distributions is considered. Standard errors of the estimated parameters are calculated from the inverse Fisher&rsquo / s information matrix and a jackknife method. Probability-probability (P-P) plots and Kolmogorov-Smirnov (K-S) goodnessof- fit tests are used to evaluate the fit of the posited distribution. A bootstrapping method is used to calculate the critical values of the K-S test statistic, which are not available.</p>

I evaluate the methods of cladistic parsimony and maximum likelihood in phylogenetic systematics by their affinity to Popper‘s degree of corroboration. My work analyzes an important recent exchange between the proponents of the two methods. Until this exchange, only advocates of cladistic parsimony have claimed a basis for their method on epistemological grounds through corroboration. Advocates of maximum likelihood, on the other hand, have based the rational justification for their method largely on statistical grounds. In Part One I outline corroboration in terms of content, severity of test and explanatory power. In Part Two I introduce the two methods. In Part Three I analyze three important debates. The first involves the appropriate probability interpretation for phylogenetics. The second is about severity of test. The third concerns explanatory power. In Part Four I conclude that corroboration can decide none of these debates, and therefore cannot decide the debate between cladistic parsimony and maximum likelihood.

In this dissertation we developed theoretical models to analyse Gaussian and multinomial distributions. The analysis is focused on classification in high dimensional feature spaces and provides a basis for dealing with issues such as data sparsity and feature selection (for Gaussian and multinomial distributions, two frequently used models for high dimensional applications). A Naïve Bayesian philosophy is followed to deal with issues associated with the curse of dimensionality. The core treatment on Gaussian and multinomial models consists of finding analytical expressions for classification error performances. Exact analytical expressions were found for calculating error rates of binary class systems with Gaussian features of arbitrary dimensionality and using any type of quadratic decision boundary (except for degenerate paraboloidal boundaries). Similarly, computationally inexpensive (and approximate) analytical error rate expressions were derived for classifiers with multinomial models. Additional issues with regards to the curse of dimensionality that are specific to multinomial models (feature sparsity) were dealt with and tested on a text-based language identification problem for all eleven official languages of South Africa. / Thesis (M.Ing. (Computer Engineering))--North-West University, Potchefstroom Campus, 2009.

I evaluate the methods of cladistic parsimony and maximum likelihood in phylogenetic systematics by their affinity to Popper‘s degree of corroboration. My work analyzes an important recent exchange between the proponents of the two methods. Until this exchange, only advocates of cladistic parsimony have claimed a basis for their method on epistemological grounds through corroboration. Advocates of maximum likelihood, on the other hand, have based the rational justification for their method largely on statistical grounds. In Part One I outline corroboration in terms of content, severity of test and explanatory power. In Part Two I introduce the two methods. In Part Three I analyze three important debates. The first involves the appropriate probability interpretation for phylogenetics. The second is about severity of test. The third concerns explanatory power. In Part Four I conclude that corroboration can decide none of these debates, and therefore cannot decide the debate between cladistic parsimony and maximum likelihood.

In this dissertation we developed theoretical models to analyse Gaussian and multinomial distributions. The analysis is focused on classification in high dimensional feature spaces and provides a basis for dealing with issues such as data sparsity and feature selection (for Gaussian and multinomial distributions, two frequently used models for high dimensional applications). A Naïve Bayesian philosophy is followed to deal with issues associated with the curse of dimensionality. The core treatment on Gaussian and multinomial models consists of finding analytical expressions for classification error performances. Exact analytical expressions were found for calculating error rates of binary class systems with Gaussian features of arbitrary dimensionality and using any type of quadratic decision boundary (except for degenerate paraboloidal boundaries). Similarly, computationally inexpensive (and approximate) analytical error rate expressions were derived for classifiers with multinomial models. Additional issues with regards to the curse of dimensionality that are specific to multinomial models (feature sparsity) were dealt with and tested on a text-based language identification problem for all eleven official languages of South Africa. / Thesis (M.Ing. (Computer Engineering))--North-West University, Potchefstroom Campus, 2009.

In this dissertation, we develop several new inference procedures that are based on randomized nomination sampling (RNS). The first problem we consider is that of constructing distribution-free confidence intervals for quantiles for finite populations. The required algorithms for computing coverage probabilities of the proposed confidence intervals are presented. The second problem we address is that of constructing nonparametric confidence intervals for infinite populations. We describe the procedures for constructing confidence intervals and compare the constructed confidence intervals in the RNS setting, both in perfect and imperfect ranking scenario, with their simple random sampling (SRS) counterparts. Recommendations for choosing the design parameters are made to achieve shorter confidence intervals than their SRS counterparts. The third problem we investigate is the construction of tolerance intervals using the RNS technique. We describe the procedures of constructing one- and two-sided RNS tolerance intervals and investigate the sample sizes required to achieve tolerance intervals which contain the determined proportions of the underlying population. We also investigate the efficiency of RNS-based tolerance intervals compared with their corresponding intervals based on SRS. A new method for estimating ranking error probabilities is proposed. The final problem we consider is that of parametric inference based on RNS. We introduce different data types associated with different situation that one might encounter using the RNS design and provide the maximum likelihood (ML) and the method of moments (MM) estimators of the parameters in two classes of distributions; proportional hazard rate (PHR) and proportional reverse hazard rate (PRHR) models.

This thesis shows how to test the fit of a data set to a number of different models, using Watson’s U2 statistic for both grouped and continuous data. While Watson’s U2 statistic was introduced for continuous data, in recent work, the statistic has been adapted for grouped data. However, when using Watson’s U2 for continuous data, the asymptotic distribution is difficult to obtain, particularly, for some skewed circular distributions that contain four or five parameters. Until now, U2 asymptotic points are worked out only for uniform distribution and the von Mises distribution among all circular distributions. We give U2 asymptotic points for the wrapped exponential distributions, and we show that U2 asymptotic points when data are grouped is usually easier to obtain for other more advanced circular distributions. In practice, all continuous data is grouped into cells whose width is decided by the accuracy of the measurement. It will be found useful to treat such data as grouped with sufficient number of cells in the examples to be analyzed. When the data are treated as grouped, asymptotic points for U2 match well with the points when the data are treated as continuous. Asymptotic theory for U2 adopted for grouped data is given in the thesis. Monte Carlo studies show that, for reasonable sample sizes, the asymptotic points will give good approximations to the p-values of the test.

In this thesis, estimation in simple linear regression model with one-fold nested error is studied. To estimate the fixed effect parameters, generalized least squares and maximum likelihood estimation procedures are reviewed. Moreover, Minimum Norm Quadratic Estimator (MINQE), Almost Unbiased Estimator (AUE) and Restricted Maximum Likelihood Estimator (REML) of variance of primary units are derived. Also, confidence intervals for the fixed effect parameters and the variance components are studied. Finally, the aforesaid estimation techniques and confidence intervals are applied to a real-life data and the results are presented