This thesis investigates robust techniques for mean-variance (MV) portfolio optimization problems under the estimation risk in mean return. We evaluate the performance of the optimal portfolios generated by the min-max robust MV portfolio optimization model. With an ellipsoidal uncertainty set based on the statistics of the sample mean estimates, minmax robust portfolios equal to the ones from the standard MV model based on the nominal mean estimates but with larger risk aversion parameters. With an interval uncertainty set for mean return, min-max robust portfolios can vary significantly with the initial data used to generate the uncertainty set. In addition, by focusing on the worst-case scenario in the mean return uncertainty set, min-max robust portfolios can be too conservative and unable to achieve a high return. Adjusting the conservatism level of min-max robust portfolios can only be achieved by excluding poor mean return scenarios from the uncertainty set, which runs counter to the principle of min-max robustness. We propose a CVaR robust MV portfolio optimization model in which the estimation risk is measured by the Conditional Value-at-Risk (CVaR). We show that, using CVaR to quantify the estimation risk in mean return, the conservatism level of CVaR robust portfolios can be more naturally adjusted by gradually including better mean return scenarios. Moreover, we compare min-max robust portfolios (with an interval uncertainty set for mean return) and CVaR robust portfolios in terms of actual frontier variation, portfolio efficiency, and portfolio diversification. Finally, a computational method based on a smoothing technique is implemented to solve the optimization problem in the CVaR robust model. We numerically show that, compared with the quadratic programming (QP) approach, the smoothing approach is more computationally efficient for computing CVaR robust portfolios.

The aim of the thesis is to identify an appropriate model in forecasting Value-at-Risk on a morevolatile period than that one from which the model is estimated. We estimate 1-day-ahead and10-days-ahead Value-at-Risk on a number of exchange rates. The Value-at-Risk estimates arebased on three models combined with three distributional assumptions of the innovations, andthe evaluations are made with Kupiec's (1995) test for unconditional coverage. The data rangesfrom January 1st 2006 through June 30th 2011. The results suggest that the GARCH(1,1) andGJR-GARCH(1,1) with normally distributed innovations are models adequately capturing theconditional variance in the series.

Methods for conditional Poisson sampling (CP-sampling) and Sampford sampling are compared and the focus is on the efficiency of the methods. The efficiency is investigated by simulation in different sampling situations. It was of interest to compare methods since new methods for both CP-sampling and Sampford sampling were introduced by Bondesson, Traat & Lundqvist in 2004. The new methods are acceptance rejection methods that use the efficient Pareto sampling method. They are found to be very efficient and useful in all situations. The list sequential methods for both CP-sampling and Sampford sampling are other methods that are found to be efficient, especially if many samples are to be generated.

In this paper, let be a sequence of k-record values from a population with common distribution function F. We will characterize the continuous (or discrete) distribution function F by the conditional expectation functions. We also study the necessary and sufficient conditions such that the conditional expectations of k-record values hold for some distribution function F. A corresponding characterization based on weak k-record values and some related characterizations are also investigated.

Participating students¡¦ association at school can enhance the students¡¦ learning, all kinds of life skills and social abilities. It¡¦s not simply a part of curriculum courses, but with more significance on cultivating the concept of being gregarious and law-abiding. Arising out of this, it comprehends positively the purpose of being well-developed education. The subject of this thesis is ¡§the Operation and Administration on the Student Arts Association at Elementary Schools¡¨, enphasizing mainly on the influences on students who have already participated the arts asssociation. From my observation on the anterior research, they emphasized more on the study of music asssociation as well as pedagogy, instead of expanding thoroughly on the operation and administration on arts asssociation. This thesis takes seven arts organizations as case studies, they separately are: Orchestra, Choir, Taiwanese Golden Loin Dance, Jazz Dance Group, Arts Association, Recorder Association, and Pottery-flute Association. The purposes of this research are, firstly, make a thorough investigation on the operation mode of the Elementary Schools¡¦ Arts Association, by the theory and method of management; secondly, in order to comprehend the operation mode of Yen-Chen Elementary School¡¦s Arts association, the interviews of each individual case are applied. Also, the interview groups can be divided into five: administrants, professors, students, graduated students, and parents of students, to the amount of twenty-two cases. This thesis draws two main conclusions.One is the suggestions for operation and administration:(1)arts association demand clearer objectives;(2)they require stronger organization;(3)they call for a flexible staff;(4)they need concise plans which should be excuted with good use of resourses;(5)with the ideas of ¡¥positive aspiration¡¦ and ¡¥encouragement leadership¡¦, arts associations aim to converge comradship;(6)arts association are supposed to recruit new members and hold the old ones;(7)The mainipulation of funds should be made public;(8)Relations among arts associations should be intensified. The other part suggests that students not only learn professions, but also cultivate appreciation. Students make friends, develop abilities and have a good time in clubs. Therefore, most of students join the same association when they go to high schools. This thesis is devided into five chapters. Chapter One is the introduction about research background and theme; Chapter Two is the literature review and theory framework; Chapter Three is the method; Chapter Four is the research process; Chapter Five is the conclusion and suggestions.

Bayesian approaches to the regression analysis for longitudinal measurements are considered. The history of measurements from a subject may convey characteristics of the subject. Hence, in a regression analysis with longitudinal measurements, the characteristics of each subject can be served as covariates, in addition to possible other covariates. Also, the longitudinal measurements may lead to complicated covariance structures within each subject and they should be modeled properly. When covariates are some unobservable characteristics of each subject, Bayesian parametric and nonparametric regressions have been considered. Although covariates are not observable directly, by virtue of longitudinal measurements, the covariates can be estimated. In this case, the measurement error problem is inevitable. Hence, a classical measurement error model is established. In the Bayesian framework, the regression function as well as all the unobservable covariates and nuisance parameters are estimated. As multiple covariates are involved, a generalized additive model is adopted, and the Bayesian backfitting algorithm is utilized for each component of the additive model. For the binary response, the logistic regression has been proposed, where the link function is estimated by the Bayesian parametric and nonparametric regressions. For the link function, introduction of latent variables make the computing fast. In the next part, each subject is assumed to be observed not at the prespecifiedtime-points. Furthermore, the time of next measurement from a subject is supposed to be dependent on the previous measurement history of the subject. For this outcome- dependent follow-up times, various modeling options and the associated analyses have been examined to investigate how outcome-dependent follow-up times affect the estimation, within the frameworks of Bayesian parametric and nonparametric regressions. Correlation structures of outcomes are based on different correlation coefficients for different subjects. First, by assuming a Poisson process for the follow- up times, regression models have been constructed. To interpret the subject-specific random effects, more flexible models are considered by introducing a latent variable for the subject-specific random effect and a survival distribution for the follow-up times. The performance of each model has been evaluated by utilizing Bayesian model assessments.

Previous research has indicated that aggression is generally detrimental to performance in the occupational domain (Campbell, 1990; James et al., 2005; Sackett, 2002; Viswesvaran et al., 1999). In certain athletic contexts, however, aggression may serve to enhance performance at the team level. For this analysis, team-level aggression is hypothesized to be positively related to team performance in basketball. Aggression in this context is defined as "the desire to inflict harm on another individual, group, or entity" (James, 2005, p.71). Both implicit (CRT-A) and explicit (NEO-PI-R) aggression were measured, and team performance was represented primarily by team scores. The data demonstrate that team-level implicit aggression is significantly and positively related to team performance, however team-level explicit aggression does not have a significant relationship with team performance.

Because of the outcomes associated with employee turnover, few areas have been researched as extensively in the field of Industrial/Organizational Psychology (Cotton&Tuttle, 1986). Correlates and consequences of employee turnover have been widely examined, though less expansive research has been dedicated to identifying (and potentially screening) turnover prone candidates. To address this concern, the comparative utility of three personality measures for predicting voluntary turnover in law enforcement was assessed. Self-report predictor measures consisted of the Minnesota Multiphasic Personality Inventory-2, the most widely used instrument in this industry, and the Personality Research Form-E. The implicit personality measure for this investigation was the Conditional Reasoning Test for Aggression (James, 1998). In line with previous research (e.g., Hough, 1998), self-report job applicant scores indicated a defensive or self-presentation bias, whereas the CRT-A did not. Normative and comparative data for all predictor measures are presented to further knowledge for the researcher and practitioner. The importance of context, in this case the influence of the economy on attrition rates, is also discussed.

Differential framing occurs when individuals with different latent motives assign qualitatively different meanings to the same attributes or events in the environment (James&Mazerolle, 2002; James&McIntyre, 1996). The implications of this phenomenon for the explanation and prediction of behavior are substantial: In perfectly logical fashions, individuals in exactly the same situation have qualitatively different experiences. In this way, differential framing mediates the relationship between motives and the behaviors that comprise traits. This dissertation tested several propositions associated with this phenomenon, and the results tentatively suggest that individuals with contrasting motives form qualitatively distinct impressions of the same organizational cultures.

Several decades of leadership research have failed to yield a personality measure that accurately predicts successful leaders (Bernus&Manis, 1985; Stogdill, 1974; Vroom&Yago, 2007; Yukl, 1989). A new implicit measure of personality, the Conditional Reasoning Test for Leadership (CRT-L), shows promise in this endeavor. This project investigated the construct and criterion-related validities of this measure. Previous research on implicit personality measures, and specifically conditional reasoning measures, has demonstrated that their relationship to their explicit measure counterparts tends to be modest or nonexistent. This was the case for the CRT-L, which had no relationship to the NEO Hostility Scale or the Motivation to Lead (MTL) Scale. As expected, the two explicit measures did have a significant and positive relationship (r = .42). The CRT-L was also effective at predicting leadership and power criteria. It had positive and significant relationships with Leadership Peer Nominations (r = .25) and Power Peer Nominations (r = .21) and was more successful in these predictions than either of the explicit measures. The results of this research provide evidence for the effectiveness of the CRT-L as a leadership measure and further validation work is encouraged.