É muito comum encontrar nas áreas agronômica e biológica experimentos cujas observações são correlacionadas. Porém, tais correlações, em tese, podem estar associadas às parcelas ou às subparcelas, dependendo do plano experimental adotado. Além disso, a metodologia de modelos lineares mistos vem sendo utilizada com mais freqüência, principalmente após os trabalhos de Searle (1988), Searle at al. (1992), Wolfinger (1993b) entre outros. O sucesso do procedimento de modelagem está fortemente associado ao exame dos efeitos aleatórios que devem permanecer no modelo e na possibilidade de se introduzir, no modelo, estruturas de variâncias e covariâncias das variáveis aleatórias que, para o modelo linear misto, podem estar inseridas no resíduo e, também, na parte aleatória associada ao fator aleatório conhecido. Nesse contexto, o Teste da Razão de Verossimilhança e o Critério de Akaike podem auxiliar na tarefa de escolha do modelo mais apropriado para análise dos dados, além de permitir verificar que escolhas de modelos inadequadas acarretam em conclusões divergentes em relação aos efeitos fixos do modelo. Com o desenvolvimento do Proc Mixed do SAS (Littel at al. 1996), utilizado neste trabalho, a análise desses experimentos, tratada pela metodologia modelos lineares mistos, tornou-se mais usual e segura. Com a finalidade de se atingir o objetivo deste trabalho, utilizaram-se dois exemplos (A e B) sobre a resposta da produtividade de três cultivares de trigo, em relação a níveis de irrigação por aspersão line-source. Foram criados e analisados 29 modelos para o Exemplo A e 16 modelos para o Exemplo B. Pôde-se verificar, para cada um dos exemplos, que as conclusões em relação aos efeitos fixos se modificaram de acordo com o modelo adotado. Notou-se, também, que o Critério de Akaike deve ser visto com cautela. Ao se comparar modelos similares entre os dois exemplos, ratificou-se a importância de se programar corretamente no Proc Mixed. Nesse contexto, conclui-se que é fundamental conduzir a análise de experimentos de forma ampla, buscando vários modelos e verificando quais têm lógica em relação ao plano experimental, evitando erros ao término da análise. / In Biology and Agronomy, experiments that produce correlated observations are often found. Theoretically, these correlations may be associated with whole-plots or subplots, according to the chosen experimental design. Also, the mixed linear model methodology is now being used much more frequently, especially after the works of Searle (1988), Searle et al. (1992) and Wolfinger (1993b), among others. The success of the modeling procedure is strongly associated with the examination of the random effects that must remain within the model and the possibility of introducing variance-covariance structures of random variables in the model. In the case of the mixed linear model, they may be included in the residual error or in the random part which is associated with the known random factor. In this context, the Likelihood Ratio Test and Akaike's Information Criterion can help in choosing the most appropriate model for data analysis. They also enable the verification of inadequate choice of models which can lead to divergent conclusions regarding the fixed effects of the model. With the development of the SAS Mixed Procedure (Little at al. 1996), which was used in this work, analysis of these experiments, conducted through the mixed linear model methodology, has become more usual and secure. In order to achieve the target of this work, two examples were utilized (A and B) involving the productivity response of three varieties of wheat, in regards to irrigation levels by line-source aspersion. Twenty-nine models for Example A and 16 models for Example B were created and analyzed. For each example, it was verified that conclusions regarding fixed effects changed according to the model adopted. It was also verified that Akaike’s Information Criterion must be regarded with caution. When comparing similar models between the two examples, the importance of correct programming in the Mixed Procedure was confirmed. In this context, it can be concluded that it is fundamental to conduct the experiment analysis in an ample manner, looking for various models and verifying which ones make sense according to the experimental plan, thus avoiding errors at analysis completion.

A técnica de tratamento de solos com cal ou cimento vem sendo empregada com sucesso na engenharia geotécnica, melhorando as características do solo, que por ser um material complexo e muito variável nem sempre satisfaz as necessidades da obra a ser realizada. As últimas pesquisas em solos tratados com cal mostram o desenvolvimento de metodologias de dosagem baseadas em critérios racionais (como a relação água/cimento para o concreto), onde a relação volume de vazios/volume de cal desempenha papel fundamental na obtenção da resistência desejada. O volume de vazios (ou a porosidade) é um fator importante nestas metodologias de dosagem, e ainda não existem técnicas que quantifiquem este fator (como o ensaio de porosimetria por injeção de mercúrio para o concreto) e também modelos que permitam o entendimento do comportamento da porosidade dessas misturas o longo do tempo de cura (como o modelo de Powers para o concreto). Assim, esta pesquisa tem como objetivo verificar a influência da quantidade de cal (Ca), da porosidade (h), do teor de umidade (w) e do tempo de cura (t), sobre a resistência à compressão simples (qu), sobre a resistência à tração (qt) e sobre a rigidez inicial (Go) de um solo siltoso estabilizado com cal (misturas caulim-cal), verificando a adequação do uso da relação vazios/cal na estimativa de qu, qt e Go. Além disso, esta pesquisa busca quantificar a porosidade dessas misturas solo-cal e também desenvolver um modelo, que permita o entendimento do comportamento da sua porosidade ao longo do tempo de cura. Para atingir os objetivos da pesquisa foram realizados ensaios de resistência à compressão simples, ensaios de resistência à tração por compressão diametral, ensaios para a medida de Go, ensaios de sucção matricial e ensaios de porosimetria por injeção de mercúrio. Os resultados dos ensaios de resistência à compressão simples, de resistência à tração e de rigidez inicial demonstram que o aumento da quantidade de Ca, a diminuição da h e o aumento do t provoca o aumento de qu, qt e Go. Sendo que, qu, qt e Go crescem linearmente com o aumento da quantidade de cal e exponencialmente com a redução da sua porosidade. Assim, verifica-se que, a relação vazios/cal (h/Cav), definida pela razão entre a porosidade da mistura compactada e o teor volumétrico de cal, demonstra ser um parâmetro adequado na estimativa de qu, qt e Go. A partir desses mesmos resultados, observase que a w também desempenha um papel fundamental na previsão de qu, qt e Go. Além disso, a partir dos ensaios de resistência à compressão simples, resistência à tração e rigidez inicial, observa-se que a existência de relações únicas e distintas no controle de qu, qt e Go em função da h, do Cav e da w mostrou-se muito eficiente para relações de dosagem. Relações entre qu, qt, Go e h/Cav também, foram muito satisfatórias. Além disso, foram realizadas análises estatísticas dos dados obtidos neste experimento, e os resultados demonstram, a partir da análise da variância, que todos os fatores controláveis escolhidos no experimento são significativos. Os resultados dos ensaios de porosímetro por intrusão de mercúrio demonstram que a porosidade diminui com o tempo de cura. Porém, o modelo de Powers não se adaptou perfeitamente na previsão da variação da porosidade das misturas caulim-cal estudadas. / The technique of treating soil with lime or cement has been used successfully in geotechnical engineering, improving the characteristics of the soil, which is a highly variable and complex material, and does not always meet the needs of the earthwork to be performed. The last researches in soils treated with lime are in the development of dosage methodologies based on rational criteria (such as water/cement ratio for concrete), where the voids/lime ratio plays a fundamental role in the assesstment of the target strength. The void volume (or porosity) is an important factor in these dosage methodologies, and there are not techniques that quantify this factor (as the test porosimetry with intrusion of mercury, for concrete) and also models that allow understanding the behavior of porosity for these mixtures in long curing times (such as the Powers’s model for concrete). Thus, this research aims to determine the influence of the amount of lime (Ca), porosity (h), moisture content (w) and curing time period (t) on the unconfined compression strength (qu), tensile strength (qt) and initial stiffness (Go) of a silty soil stabilized with lime (kaolin-lime mixtures), checking the suitability of the use of voids/lime ratio in estimating qu, qt and Go. Besides, this research aims to quantify the porosity of these soil-lime mixtures and also adjusting a model that allows understanding the behavior of their porosity during the curing time. For that, a number of unconfined compression tests, splitting tensile tests, the measurement of Go, measurement of matric suction and porosimetry with intrusion of mercury tests were carried out in present work. The results of unconfined compression strength, tensile strength and initial stiffness show that increasing the amount of Ca, decreasing of h and increasing of t, causes increased of qu, qt and Go. Further, qu, qt and Go grow linearly with the increased amount of lime and exponentially with reducing its porosity. The voids/lime ratio, defined as the ratio of the compacted mixture porosity and the lime volumetric content, adjusted by an exponent, proves to be an appropriate parameter to estimate the qu, qt and Go. From these results, it is observed that the w also plays a fundamental parameter in predicting the qu, qt and Go. Moreover, it is noted that the existence of distinct and unique relationships in the control of qu, qt and Go according to h, Cav and w proved to be very efficient for dosage relationships. Relations between qu, qt, Go and h/Cav were very satisfactory too. Furthermore, statistical analyzes were performed of the results obtained in this experiment, and results demonstrate, through analysis of variance, that all controllable factors chosen in the experiment are significant. The results of test porosimetry with intrusion of mercury show that the porosity decreases with increasing curing time. However, the Powers’s model has not adapted perfectly to predict the variation of the porosity of kaolin-lime mixtures studied.

The purpose of the present study was to examine the effect of varying the amount of participant engagement on alcohol expectancy and drinking outcomes during a social/sexual expectancy challenge based on Darkes and Goldman's (1993, 1998) protocol. This study was also intended to provide a test of the efficacy of administering an alcohol/placebo expectancy challenge outside of a live drinking scenario through video presented as part of a computerized intervention. One hundred fifty-eight male participants across three sites were randomized into a no-intervention control group that received non alcohol-related information in a minimally interactive computerized format, a low-level engagement experimental group that received minimally interactive computerized expectancy-related information, and a high-level engagement experimental group that received the same expectancy-related information presented in a more interactive computerized format that included games and audiovisual elements such as video clips, graphics, live narrations, and music. It was hypothesized that high-level engagement participants would report being more engaged in their computerized program and demonstrate greater decreases in social/sexual alcohol expectancies and drinking levels relative to control and low-level engagement participants. Results indicated that while high-level engagement participants reported being more engaged in their interventions, none of the groups exhibited changes in the alcohol expectancies measured. In addition, all three groups experienced significant but comparable decreases in drinking levels. Exploratory follow-up analyses were also conducted to provide suggestions for future directions.

National cultures play a critical role in effective leadership and organisational success in international businesses. Contemporary organisations are therefore challenged by cross-cultural leadership needs to address increasing diversity. In the past decades, there has been a growing interest in studies of non-Western leadership. Recent papers have focused particularly on leadership in the Asia-Pacific region. This paper compares leadership in China and New Zealand, based on the data collected by using the GLOBE (Global Leadership and Organisational Behaviour Effectiveness) project leadership questionnaire. In addition, the paper reviews the main findings of previous research in order to investigate the similarities and differences in preferred leadership behaviours/characteristics in both cultures, while attempting to examine the consistency of current findings against the previous GLOBE studies. Significant differences of leadership behaviours between China and New Zealand have been identified. This paper generally supports the findings of previous GLOBE studies that charismatic/value-based leadership is the universally contributing factor towards outstanding leadership and self-protective leadership is the universally negative factor against leadership excellence. Findings in this paper may contribute to further understanding of leadership practice in these two countries. Future research should be undertaken to examine the effects of cultural differences on organisational practice by a more comprehensive research method.

Identification of nonlinear dynamical models of a black box nature involves both structure decisions (i.e., which regressors to use and the selection of a regressor function), and the estimation of the parameters involved. The typical approach in system identification is often a mix of all these steps, which for example means that the selection of regressors is based on the fits that is achieved for different choices. Alternatively one could then interpret the regressor selection as based on hypothesis tests (F-tests) at a certain confidence level that depends on the data. It would in many cases be desirable to decide which regressors to use, independently of the other steps. A survey of regressor selection methods used for linear regression and nonlinear identification problems is given. In this thesis we investigate what the well known method of analysis of variance (ANOVA) can offer for this problem. System identification applications violate many of the ideal conditions for which ANOVA was designed and we study how the method performs under such non-ideal conditions. It turns out that ANOVA gives better and more homogeneous results compared to several other regressor selection methods. Some practical aspects are discussed, especially how to categorise the data set for the use of ANOVA, and whether to balance the data set used for structure identification or not. An ANOVA-based method, Test of Interactions using Layout for Intermixed ANOVA (TILIA), for regressor selection in typical system identification problems with many candidate regressors is developed and tested with good performance on a variety of simulated and measured data sets. Typical system identification applications of ANOVA, such as guiding the choice of linear terms in the regression vector and the choice of regime variables in local linear models, are investigated. It is also shown that the ANOVA problem can be recast as an optimisation problem. Two modified, convex versions of the ANOVA optimisation problem are then proposed, and it turns out that they are closely related to the nn-garrote and wavelet shrinkage methods, respectively. In the case of balanced data, it is also shown that the methods have a nice orthogonality property in the sense that different groups of parameters can be computed independently.

According to the change of Taiwan economic structure, Taiwan is gradually moving into an aging society and more and more population in Taiwan will suffer from chronic disease in the near future. Hypertension is not only the first cause of death among the top ten COD in Taiwan, but is also the major source of other four types of disease in the top ten COD list. Market size for hypertension treatment has enormous potential thus all multi-national pharmaceutical companies seize for opportunities in it. Hypertension treatment research and development are always a priority for pharmaceutical companies and medical research centers, however, new compound takes a long time to develop and patent protection period is getting shorter, therefore, how to increase physician acceptance of a new compound becomes the critical success factor of the new launch product for pharmaceutical companies. This research is an attempt to study the relationship between ¡§Current Prescription¡¨, ¡§New Drug Usage¡¨ and ¡§Environmental Factors-Medical Provider¡¨ and ultimately to find the vital factors which would influence the ¡§Willingness¡¨ of the physicians to prescribe product X. One aim of this research is also to understand the potential factors which would affect the physician acceptance of a new mechanism drug. Based on the Multiple Regression Analysis, Analysis of Variance and Factor Analysis, the research results are as follows: 1. The effect of ¡§Current Prescription¡¨ on ¡§Willingness to Prescribe New Product X¡¨ is limited, and factors such as ¡§Satisfaction on Current Prescription¡¨ and ¡§Current Prescription¡¨ do not influence the ¡§Willingness to Prescribe New Product X¡¨. Only the satisfaction of BLOPRESS and PLENDIL and current prescription percentage on ADALOT OROS and CO-DIOVAN show a significant affect. So the research hypothesis H-1 and H1-3 are partially supported. AS far as ¡§Level of Fame¡¨ of the product shows a significant influence on the ¡§Willingness to Prescribe New Product X¡¨, so the research hypothesis H1-2 is fully supported. 2. Relationship between ¡§New Drug Usage¡¨ and ¡§Willingness to Prescribe New Product X¡¨ is not all significant. ¡§External Information¡¨, ¡§Physician Specialty¡¨ and factors a physician will consider when prescribing did not have a large impact on whether the physician will prescribe the new drug or not. Only the familiarity on ¡§Renin system is an import treatment indicator on blood pressure control and organ protection¡¨, ¡§External Information Orientated¡¨, and ¡§Market Follower¡¨ type of physicians and ¡§Will still use broadly to treat, even despite the high cost of new drugs¡¨ and ¡§New product X will always be part of the prescription in combination therapy¡¨ show significant impact. Thus hypothesis H2-1, H2-3 & H2-4 are partial supported. ¡¨Type of Patient¡¨ shows no difference in effect physician¡¦s willingness to prescribe new product X, so the hypothesis H2-2 are not supported. 3. ¡§Level of Fame¡¨ just has partial significance effect to physician¡¦s ¡§Current Prescription¡¨. ¡§Level of fame¡¨ for CCB, ARB and ACEI have significant effect to the physician¡¦s ¡§Current Prescription¡¨, but other drugs like ALPHA BLOCKER, BETA BLOCKER & OTHERS have no effect to the physician¡¦s ¡§Current Prescription¡¨, so the research hypothesis is partially supported. 4. ¡§Environment Factors¡ÐMedical Provider¡¨ has limited effect to physician¡¦s ¡§Current Prescription¡¨. Among all the physician specialties, only Cardiovascular, Nephrology and Meta/Endo physicians show a difference in ¡§Current Prescription¡¨ among all specialties. ¡§Type of Hospital¡¨ shows no effect on ¡§Current Prescription¡¨, thus the research hypothesis is not supported. According to the findings above, pharmaceutical companies should increase call frequency from sales representatives to the physicians; increased call frequency will effectively increase the brand awareness of the new launch products and also to increase the new product prescription rate. Improvement of Sales Training and after sales service can also be a key area for improved for the pharmaceutical companies to improve the effectiveness and efficiency for product promotion. In addition, product efficacy and treatment differentiation messages should also be customized for physicians in different specialties and in different types of hospitals; increased frequency and customized messages will ensure the success of the new product launch.

Draper and Guttman (1997) show that for basic 2^{k-p}designs, p >= 0, k-p replicates of blocks designs of size two are needed to estimate all the usual (estimable) effects. In Chapter 1, we provide an algebraic formal proof for the two-level blocks designs results and present results applicable to the general case; that is, for the case of s^{k} factorial (p=0) or s^{k-p}ractional factorial (p>0) designs in s^{b} blocks, where 0<b<k-p, at least (k-p)/(k-p-b) replicates are needed to clear up all possible effects. Through the theoretical development presented in this work, it can provide a clearer view on why those results would hold. We will also discuss the estimation equations given in Draper and Guttman (1997). In Chapter 2, we present a methodology for analyzing the variability of the radiated emission testings of electronic, elecommunication and information technology equipment based on a modified analysis of variance (ANOVA), with polynomial regression analysis. In our study, three electronic products; modem, monitor and notebook bought from the market are tested. Through the experiment, we show that the international standard fails to provide a methodology which gives control limits for EMC when the electronic products in question are produced. We feel that an improved EMC control procedure presented here can better meet the needs of radiated emission control.

While evaluating the depth of corrosive defect through Remote Field Eddy Current (RFEC) Testing technology, the researcher tried to investigate the signals of supporting plate which may produce variations by the thickness of supporting plate, diameter of tube hole in supporting plate, the value of crevice between tube and tube hole and tube wall thickness. Errors of evaluation of defect depth may consequently be identified or measured by the variations of the support plate signal. This study explores the effects of above four factors by experiments and an analysis of variance in statistics. By the analysis of experimental results, the researcher found the four influential factors would cause angle deviation of supporting plate signals. Except the factor of tube wall thickness, the deviation is not so substantial that the difference of evaluation in depth was consequently fallen into an acceptable range of engineering practices. When utilizing the remote field eddy current testing technique to test inservice tubes, the researcher found the thickness of inservice tube is normally different from the specified thickness of a standard tube. This variation consequently resulting in an evaluation curve produced by a standard tube may not lead to proper assessment of defects in inservice tube. To deal with the problem many researchers used a frequency compensation method to compensate for the evaluation error contributed by the variation of tube thickness on the basis of the standard ASTM E2096-00(2000). But the use of the frequency compensation method did not measure the inservice tube thickness and thus produced a drawback of consuming much time for the adjustment steps of compensation. Therefore, a mathematical compensation method was introduced in this study for the compensation of thickness variation derived from skin depth theory. This method in the present experiment is proven to be both feasible and reasonable through the derivation of the methodology. Generally, this study aims to apply the mathematical compensation method to overcome the drawback of frequency compensation method and to solve the problem of the difficulty in measuring inservice tube wall thickness in heat exchanger bundles.

The pharmacy industry in Taiwan has produced a succession of great changes in recent years, for example: The National Health Insurance, the division of medical treatment & medication, Join the World Trade Organization(WTO) which were putting into practice. However, the competent authority which protects the medicine price and nuclear price policying, bringing a large impact to the pharmaceutical factory. So the promotion sales are effected first. So, for be able to improve the sales¡¦ performance & understand the releationship of human resources management activities & sales¡¦ performance, this research attempts to take one pharmacy company for an example, and using Multiple Regression Anylysis to research the effect of Selection, Traing & Education, Salary & Welfare, Position Alternation & Employees Participating to Sales Amount, Achievement Rat of Sales, Leave Application & Customer Developing. Besides this, it also using the Analysis of Variance to understand the difference of sales¡¦ performance under the situation of Cost Reducing & Human Resource Promoting. So according to the Multiple Regression Anylysis & Analysis of Variance, this research get the results as follow: 1.Traing & Education, Salary & Welfare, Employees Participating, the three Human Resource Management Activities have the significance effect to Sales¡¦ Performance. So the hypothesis H-1, H1-3 & H1-5 are all supported. 2.Selection also has the significance effect to Sales¡¦ Performance, but not all of factors can anticipate the Sales¡¦ Performance. So in accordance with the hypothesis H1-1 just present a partical supported result. 3.Position Alternation just has the significance effect to Sales Amount, Leave Application & Customer Developing, but not has the significance effect to Achievement Rat of Sales. So in accordance with the hypothesis H1-4 just present a partical supported result. 4.Under the difference Type of Human Resource Management, The sales¡¦s performance not only has the significance effect to Sales Amount & Customer Developing, but also has the higher performance. However, there are no significance about this difference. So in accordance with the hypothesis H2-1 & H2-2 just present a partical supported result. According to above-mentioned results, if we want to enhance some of sales¡¦ performances, we should enforance some of the optional human resources activities. This research also discover that: The higher salary & welfare structure & The type of human resources promoting, which have the significance effect to increase the sales¡¦ performance.

In this thesis, we first show that the performance of ranking and selection (R&S) procedures in steady-state simulations depends highly on the quality of the variance estimates that are used. We study the performance of R&S procedures using three variance estimators --- overlapping area, overlapping Cramer--von Mises, and overlapping modified jackknifed Durbin--Watson estimators --- that show better long-run performance than other estimators previously used in conjunction with R&S procedures for steady-state simulations. We devote additional study to the development of the new overlapping modified jackknifed Durbin--Watson estimator and demonstrate some of its useful properties. Next, we consider the problem of finding the best simulated system under a primary performance measure, while also satisfying stochastic constraints on secondary performance measures, known as constrained ranking and selection. We first present a new framework that allows certain systems to become dormant, halting sampling for those systems as the procedure continues. We also develop general procedures for constrained R&S that guarantee a nominal probability of correct selection, under any number of constraints and correlation across systems. In addition, we address new topics critical to efficiency of the these procedures, namely the allocation of error between feasibility check and selection, the use of common random numbers, and the cost of switching between simulated systems.