A multivariate parallelogram and its application to multivariate trimmed means

This paper introduces a multivariate parallelogram that can play the role of the univariate quantile in the location model, and uses it to define a multivariate trimmed mean. It assesses the asymptotic efficiency of the proposed multivariate trimmed mean by its asymptotic variance and by Monte Carlo simulation.     

A TWO-STAGE UNRELATED RANDOMISED RESPONSE PROCEDURE

In a sample survey, questions requiring personal or controversial assertions often give rise to resistance. A randomised response procedure can be used to help the researcher gather accurate data in this case. This paper describes a new two-stage unrelated randomised response procedure that combines the use of two randomisation devices (Mangat & Singh, 1990) and an unrelated question (Horwitz et al. 1967). It examines the situation where the respondents are not completely truthful in their answers. The efficiency of this new method is compared with the original one-stage procedure proposed by Horwitz et al. (1967), and guidelines for choosing the values of different parameters for the procedures are provided. Results from an empirical study which examines the efficiency and feasibility of the proposed method are given.     

Lek mating system and its impact on male annihilation technique

Summary Simulation models were presented which describe the matings inside a lek. The size of the lek was assumed to follow a poisson distribution, the dominance factor of each of the males was drawn from a gamma distribution, and the mating probability for each male was determined by its mating effectiveness raised to female mating factor divided by the sum of all the mating effectivenesses raised to this power. The mating probatility for each male and the actual matings were determined by Monte Carlo simulation. Based on the simulation data, the effects of female mating factor and size of the lek on the frequency of actual matings accomplished by the males inside the lek were discussed. In the case of male annihilation, the mean number of effective matings per male in the lek, calculated from the frequency of actual matings weighted by the effective mating ratio, was used to evaluate the efficiency of male annihilation method.     

Monitoring Nonlinear Profiles with Random Effects by Nonparametric Regression

The monitoring of process/product profiles is presently a growing and promising area of research in statistical process control. This study is aimed at developing monitoring schemes for nonlinear profiles with random effects. We utilize the technique of principal components analysis to analyze the covariance structure of the profiles and propose monitoring schemes based on principal component (PC) scores. The number of the PC scores used in constructing control charts is crucial to the detecting power. In the Phase I analysis of historical data, due to the dependency of the PC-scores, we adopt the usual Hotelling T ² chart to check the stability. For Phase II monitoring, we study individual PC-score control charts, a combined chart scheme that combines all the PC-score charts, and a T ² chart. Although an individual PC-score chart may be perfect for monitoring a particular mode of variation, a chart that can detect general shifts, such as the T ² chart and the combined chart scheme, is more feasible in practice. The performances of the schemes under study are evaluated in terms of the average run length.     

Do Rural Students Really Perform Worse than Urban Students Do? Empirical Evidence from a University Entrance Program in Taiwan

Despite a major expansion in the number of students in higher education, students from rural areas continue to be underrepresented at selective universities. To reduce the urban‐rural imbalance of entry to selective universities, institutions in many countries of the world have implemented admission policies favoring rural students. Previous evidence has shown that rural students have lower academic performance than their urban peers, which leads to concern that rural students will reduce the academic excellence of selective universities. Using the introduction of a university admission policy favoring rural students in Taiwan and a unique administrative data set of students' academic records, we compare the academic performance of students from rural areas with that of their urban counterparts during their college years. The results show that rural students have consistently outperformed urban students in semester grade point averages and class rank percentile across time; however, the differences in academic performance are mainly attributed to the rural students' initial outperformance in the first semester of the freshman year.     

Impact of lek mating on the sterile insect technique: A modeling study

Summary Modelling studies are presented which describe the effect of lek mating on the control of a wild population by sterile male release. The mixed leks are assumed to follow a Poisson-binomial distribution and the system includes three parts: territory defense, matings inside a lek and matings outside a lek. The effects of parameters on the hatchability are discussed. Among the parameters, sterile type effect (W _s ), female choice (f _s ) and mating competitiveness (C _m ) are the most important. The application to determining the effects of sterile male release and on the proportion of sterile males required for eradication are also discussed.     

A note on mse coverage intervals in a partial spline model

A partial spline model is used to estimate an unknown function which is smooth except for some break points. Assuming the break points are known, a Generalized Cross-Validated smoothing spline estimation method is proposed. Some interval estimation methods for the magnitude of the discontinuities based on the mean square error are introduced and investigated.     

Behavioral processes and egg-laying decisions of the bean weevil, Callosobruchus maculatus

Empirical studies of behavioral processes were designed in a seed beetle, Callosobruchus maculatus, to distinguish between two alternative tactical models, namely, comparison tactics and threshold tactics of seed choice, and also to deduce the behavioral rule of its oviposition decisions. Search sequences and the oviposition process of the female bruchid beetle were observed and recorded. Analyzing data of the search sequence showed that the female bruchid beetle repeatedly encountered seeds randomly and tended to oviposit on a newly encountered seed in an ovposition bout. These results contradict the usual comparison models, which predict that the female will return to choose any one of the previously encountered seeds after sampling. In addition, the oviposition decision (rejection or acceptance of an encountered seed to oviposit) was analyzed with a generalized linear modeling (GLIM) technique. Modeling showed that the probability of accepting a seed with different numbers of eggs on it changed during her egg-laying process. This evidence supports the idea that the female is using a threshold tactic and that her acceptance threshold is being adjusted by experience gained during the egg-laying process. The analysis of statistical modeling also showed that both the time since the last oviposition and the number of eggs which had been laid by the female had a significant effect on the probability of accepting seeds with different numbers of eggs. Therefore, a time measuring system and the physiological state variable, e.g., eggload of the female, should be included in the behavioral rule to explore how the female makes her decision in the egg-laying process and to examine the importance of choice behavior as a component of selection and adaptation. Received: May 8, 1998 / Accepted: September 20, 1999