Many minimum distance estimators have the potential to provide parameter estimates which are both robust and efficient and yet, despite these highly desirable theoretical properties, they are rarely used in practice. This is because the performance of these estimators is rarely guaranteed per se but obtained by placing a suitable value on some tuning parameter. Hence there is a risk involved in implementing these methods because if the value chosen for the tuning parameter is inappropriate for the data to which the method is applied, the resulting estimators may not have the desired theoretical properties and could even perform less well than one of the simpler, more widely used alternatives. There are currently no data-based methods available for deciding what value one should place on these tuning parameters hence the primary aim of this research is to develop an objective way of selecting values for the tuning parameters in minimum distance estimators so that the full potential of these estimators might be realised. This new method was initially developed to optimise the performance of the density power divergence estimator, which was proposed by Basu, Harris, Hjort and Jones [3]. The results were very promising so the method was then applied to two other minimum distance estimators and the results compared.

Ever since the pioneering work of Parzen the mean square error( MSE) and its integrated form (MISE) have been used as the error criteria in choosing the bandwidth matrix for multivariate kernel density estimation. More recently other criteria have been advocated as competitors to the MISE, such as the mean absolute error. In this study we define a weighted version of the Hellinger distance for multivariate densities and show that it has an asymptotic form, which is one-fourth the asymptotic MISE under weak smoothness conditions on the multivariate density f. In addition the proposed criteria give rise to a new data-dependent bandwidth matrix selector. The performance of the new data-dependent bandwidth matrix selector is compared with other well known bandwidth matrix selectors such as the least squared cross validation (LSCV) and the plug-in (HPI) through simulation. We derived a closed form formula for the mean Hellinger distance (MHD) in the univariate case. We also compared via simulation mean weighted Hellinger distance (MWHD) and the asymptotic MWHD, and the MISE and the asymptotic MISE for both univariate and bivariate cases for various densities and sample sizes.

Robust estimation of statistical parameters is traditionally believed to exist in a trade space between robustness and efficiency. This thesis examines the Minimum Hellinger Distance Estimator (MHDE), which is known to have desirable robustness properties as well as desirable efficiency properties. This thesis confirms that the MHDE is simultaneously robust against outliers and asymptotically efficient in the univariate location case. Robustness results are then extended to the case of simple linear regression, where the MHDE is shown empirically to have a breakdown point of 50%. A geometric algorithm for solution of the MHDE is developed and implemented. The algorithm utilizes the Riemannian manifold properties of the statistical model to achieve an algorithmic speedup. The MHDE is then applied to an illustrative problem in power system state estimation. The power system is modeled as a structured linear regression problem via a linearized direct current model; robustness results in this context have been investigated and future research areas have been identified from both a statistical perspective as well as an algorithm design standpoint. / Master of Science

Master of Science / Department of Statistics / Weixin Yao / In this report, we introduce the minimum Hellinger distance (MHD) estimation method and review its history. We examine the use of Hellinger distance to obtain a new efficient and robust estimator for a class of semiparametric mixture models where one component has known distribution while the other component and the mixing proportion are unknown. Such semiparametric mixture models have been used in biology and the sequential clustering algorithm. Our new estimate is based on the MHD, which has been shown to have good efficiency and robustness properties. We use simulation studies to illustrate the finite sample performance of the proposed estimate and compare it to some other existing approaches. Our empirical studies demonstrate that the proposed minimum Hellinger distance estimator (MHDE) works at least as well as some existing estimators for most of the examples considered and outperforms the existing estimators when the data are under contamination. A real data set application is also provided to illustrate the effectiveness of our proposed methodology.

In this dissertation, robust and efficient alternatives to quasi-likelihood estimation and likelihood ratio tests are developed for discrete generalized linear models. The estimation method considered is a penalized minimum Hellinger distance procedure that generalizes a procedure developed by Harris and Basu for estimating parameters of a single discrete probability distribution from a random sample. A bootstrap algorithm is proposed to select the weight of the penalty term. Simulations are carried out to compare the new estimators with quasi-likelihood estimation. The robustness of the estimation procedure is demonstrated by simulation work and by Hapel's α-influence curve. Penalized minimum Hellinger deviance tests for goodness-of-fit and for testing nested linear hypotheses are proposed and simulated. A nonparametric bootstrap algorithm is proposed to obtain critical values for the testing procedure.

No description available.

The thesis focuses on the development of statistical theory in experimental design with applications in global optimization. It consists of four parts. In the first part, a criterion of design efficiency, under model uncertainty, is studied with reference to possibly nonregular fractions of general factorials. The results are followed by a numerical study and the findings are compared with those based on other design criteria. In the second part, optimal designs are dentified using Bayesian methods. This work is linked with response surface methodology where the first step is to perform factor screening, followed by response surface exploration using different experiment plans. A Bayesian analysis approach is used that aims to achieve both goals using one experiment design. In addition we use a Bayesian design criterion, based on the priors for the analysis approach. This creates an integrated design and analysis framework. To distinguish between competing models, the HD criterion is used, which is based on the pairwise Hellinger distance between predictive densities. Mixed-level fractional factorial designs are commonly used in practice but its aliasing relations have not been studied in full rigor. These designs take the form of a product array. Aliasing patterns of mixed level factorial designs are discussed in the third part. In the fourth part, design of experiment ideas are used to introduce a new global optimization technique called SELC (Sequential Elimination of Level Combinations), which is motivated by genetic algorithms but finds the optimum faster. The two key features of the SELC algorithm, namely, forbidden array and weighted mutation, enhance the performance of the search procedure. Illustration is given with the optimization of three functions, one of which is from Shekel's family. A real example on compound optimization is also given.

O objetivo deste trabalho é investigar o desempenho de estimadores baseados em momentos das famílias verossimilhança empírica generalizada (GEL) e mínimo contraste generalizado (GMC) na estimação de modelos de equilíbrio geral dinâmico e estocástico (DSGE), com enfoque na análise de robustez sob má-especificação, recorrente neste tipo de modelo. Como benchmark utilizamos método do momentos generalizado (GMM), máxima verossimilhança (ML) e inferência bayesiana (BI). Trabalhamos com um modelo de ciclos reais de negócios (RBC) que pode ser considerado o núcleo de modelos DSGE, apresenta dificuldades similares e facilita a análise dos resultados devido ao menor número de parâmetros. Verificamos por meio de experimentos de Monte Carlo se os estimadores estudados entregam resultados satisfatórios em termos de média, mediana, viés, erro quadrático médio, erro absoluto médio e verificamos a distribuição das estimativas geradas por cada estimador. Dentre os principais resultados estão: (i) o estimador verossimilhança empírica (EL) - assim como sua versão com condições de momento suavizadas (SEL) - e a inferência bayesiana (BI) foram, nesta ordem, os que obtiveram os melhores desempenhos, inclusive nos casos de especificação incorreta; (ii) os estimadores continous updating empirical likelihood (CUE), mínima distância de Hellinger (HD), exponential tilting (ET) e suas versões suavizadas apresentaram desempenho comparativo intermediário; (iii) o desempenho dos estimadores exponentially tilted empirical likelihood (ETEL), exponential tilting Hellinger distance (ETHD) e suas versões suavizadas foi bastante comprometido pela ocorrência de estimativas atípicas; (iv) as versões com e sem suavização das condições de momento dos estimadores das famílias GEL/GMC apresentaram desempenhos muito similares; (v) os estimadores GMM, principalmente no caso sobreidentificado, e ML apresentaram desempenhos consideravelmente abaixo de boa parte de seus concorrentes / The objective of this work is to investigate the performance of moment-based estimators of the generalized empirical likelihood (GEL) and generalized minimum contrast (GMC) families in the estimation of dynamic stochastic general equilibrium (DSGE) models, focusing on the robustness analysis under misspecification, recurrent in this model. As benchmark we used generalized method of moments (GMM), maximum likelihood (ML) and Bayesian inference (BI). We work with a real business cycle (RBC) model that can be considered the core of DSGE models, presents similar difficulties and facilitates the analysis of results due to lower number of parameters. We verified, via Monte Carlo experiments, whether the studied estimators presented satisfactory results in terms of mean, median, bias, mean square error, mean absolute error and we verified the distribution of the estimates generated by each estimator. Among the main results are: (i) empirical likelihood (EL) estimator - as well as its version with smoothed moment conditions (SEL) - and Bayesian inference (BI) were, in that order, the ones that obtained the best performances, even in misspecification cases; (ii) continuous updating empirical likelihood (CUE), minimum Hellinger distance (HD), exponential tilting (ET) estimators and their smoothed versions exhibit intermediate comparative performance; (iii) performance of exponentially tilted empirical likelihood (ETEL), exponential tilting Hellinger distance (ETHD) and its smoothed versions was seriously compromised by atypical estimates; (iv) smoothed and non-smoothed GEL/GMC estimators exhibit very similar performances; (v) GMM, especially in the over-identified case, and ML estimators had lower performance than their competitors

O objetivo deste trabalho é investigar o desempenho de estimadores baseados em momentos das famílias verossimilhança empírica generalizada (GEL) e mínimo contraste generalizado (GMC) na estimação de modelos de equilíbrio geral dinâmico e estocástico (DSGE), com enfoque na análise de robustez sob má-especificação, recorrente neste tipo de modelo. Como benchmark utilizamos método do momentos generalizado (GMM), máxima verossimilhança (ML) e inferência bayesiana (BI). Trabalhamos com um modelo de ciclos reais de negócios (RBC) que pode ser considerado o núcleo de modelos DSGE, apresenta dificuldades similares e facilita a análise dos resultados devido ao menor número de parâmetros. Verificamos por meio de experimentos de Monte Carlo se os estimadores estudados entregam resultados satisfatórios em termos de média, mediana, viés, erro quadrático médio, erro absoluto médio e verificamos a distribuição das estimativas geradas por cada estimador. Dentre os principais resultados estão: (i) o estimador verossimilhança empírica (EL) - assim como sua versão com condições de momento suavizadas (SEL) - e a inferência bayesiana (BI) foram, nesta ordem, os que obtiveram os melhores desempenhos, inclusive nos casos de especificação incorreta; (ii) os estimadores continous updating empirical likelihood (CUE), mínima distância de Hellinger (HD), exponential tilting (ET) e suas versões suavizadas apresentaram desempenho comparativo intermediário; (iii) o desempenho dos estimadores exponentially tilted empirical likelihood (ETEL), exponential tilting Hellinger distance (ETHD) e suas versões suavizadas foi bastante comprometido pela ocorrência de estimativas atípicas; (iv) as versões com e sem suavização das condições de momento dos estimadores das famílias GEL/GMC apresentaram desempenhos muito similares; (v) os estimadores GMM, principalmente no caso sobreidentificado, e ML apresentaram desempenhos consideravelmente abaixo de boa parte de seus concorrentes / The objective of this work is to investigate the performance of moment-based estimators of the generalized empirical likelihood (GEL) and generalized minimum contrast (GMC) families in the estimation of dynamic stochastic general equilibrium (DSGE) models, focusing on the robustness analysis under misspecification, recurrent in this model. As benchmark we used generalized method of moments (GMM), maximum likelihood (ML) and Bayesian inference (BI). We work with a real business cycle (RBC) model that can be considered the core of DSGE models, presents similar difficulties and facilitates the analysis of results due to lower number of parameters. We verified, via Monte Carlo experiments, whether the studied estimators presented satisfactory results in terms of mean, median, bias, mean square error, mean absolute error and we verified the distribution of the estimates generated by each estimator. Among the main results are: (i) empirical likelihood (EL) estimator - as well as its version with smoothed moment conditions (SEL) - and Bayesian inference (BI) were, in that order, the ones that obtained the best performances, even in misspecification cases; (ii) continuous updating empirical likelihood (CUE), minimum Hellinger distance (HD), exponential tilting (ET) estimators and their smoothed versions exhibit intermediate comparative performance; (iii) performance of exponentially tilted empirical likelihood (ETEL), exponential tilting Hellinger distance (ETHD) and its smoothed versions was seriously compromised by atypical estimates; (iv) smoothed and non-smoothed GEL/GMC estimators exhibit very similar performances; (v) GMM, especially in the over-identified case, and ML estimators had lower performance than their competitors

No description available.