Thesis (M.S.)--Virginia Polytechnic Institute and State University, 1990. / Vita. Abstract. Includes bibliographical references (leaves 121-130). Also available via the Internet.

No description available.

No description available.

No description available.

Typescript (photocopy). / Digitized by Kansas Correctional Industries

Thesis (Ph. D.)--Ohio State University, 2003. / Title from first page of PDF file. Document formatted into pages; contains xvii, 104 p.; also includes graphics (some col.). Includes abstract and vita. Advisor: Andy Ward, Dept.of Food, Agricultural, and Biological Engineering. Includes bibliographical references (p. 100-104).

D. Tech. Mathematical technology Tshwane University of Technology 2013. / Proposes a tool or technique that greatly reduces the mathematical difculties encountered when investigating the stability of predator-prey models. The technique replaces the need to deal with complex Lienard systems, Dulac and Lyapunov functions, and the required mathematics thus becomes accessible to researchers in a wide range of disciplines.

published_or_final_version / Mechanical Engineering / Master / Master of Philosophy

No description available.

This thesis presents methodological contributions to the modeling of regimes in the time or space domain of economic data by introducing a number of algorithms from engineering applications and substantially modifying them so that can be used in economic applications. The objective is twofold: to estimate the parameters of such models, and to identify the corresponding boundaries between regimes. The models used belong to the class of Finite Mixture Models and their natural extensions for the case of dependent data, Hidden Markov Models (see McLachlan and Peel 2000). Mixture models are extremely useful in the modeling of heterogeneity in a cluster analysis context; the components of the mixtures, or the states, will correspond to the different latent groups, e.g. homogeneous regions such as the housing submarkets or regimes in the case of stock market returns. / The thesis discusses issues of alternative estimation algorithms that provide larger model flexibility in capturing the underlying data dynamics, and of procedures that allow the selection of the number of the regimes in the data. / The first part introduces a model of spatial association for housing markets, which is approached in the context of spatial heterogeneity. A Hedonic Price Index model is considered, i.e. a model where the price of the dwelling is determined by its structural and neighborhood characteristics. Remaining spatial heterogeneity is modeled as a Finite Mixture Model for the residuals of the Hedonic Index. The Finite Mixture Model is estimated using the Figueiredo and Jain (2002) approach. The overall ability of the model to identify spatial heterogeneity is evaluated through a set of simulations. The model was applied to Los Angeles County housing prices data for the year 2002. The statistically identified number of submarkets, after taking into account the dwellings' structural characteristics, are found to be considerably fewer than the ones imposed either by geographical or administrative boundaries, thus making it more suitable for mass assessment applications. / The second part of the thesis introduces a Duration Hidden Markov Model to represent regime switches in the stock market; the duration of each state of the Markov Chain is explicitly modeled as a random variable that depends on a set of exogenous variables. Therefore, the model not only allows the endogenous determination of the different regimes but also estimates the effect of the explanatory variables on the regimes' durations. The model is estimated on NYSE returns using the short-term interest rate and the interest rate spread as exogenous variables. The estimation results coincide with existing findings in the literature, in terms of regimes' characteristics, and are compatible with basic economic intuition, in terms of the effect of the exogenous variables on regimes' durations. / The final part of the thesis considers a Hidden Markov Model (HMM) approach in order to perform the task of detecting structural breaks, which are defined as the data points where the underlying Markov Chain switches from one state to another: A new methodology is proposed in order to estimate all aspects of the model: number of regimes, parameters of the model corresponding to each regime, and the locations of regime switches. One of the main advantages of the proposed methodology is that it allows for different model specifications across regimes. The performance of the overall procedure, denoted IMI by the initials of the component algorithms is validated by two sets of simulations: one in which only the parameters are permitted to differ across regimes, and one that also permits differences in the functional forms. The IMI method performs very well across all specifications in both sets of simulations.