Component selection norms for principal components regression

Multicollinearity or near exact linear dependence among the vectors of regressor variables in a multiple linear regression analysis can have important effects on the quality of least squares parameter estimates. One frequently suggested approach for these problems is principal components regression. This paper investigates alternative variable selection procedures and their implications for such an analysis.     

Exact fisher information for censored samples and the extended hazard rate functions

The question of how much information is contained in an ordered observation was studied by Tukey (1964) in terms of a linear sensitivity measure. This paper deals with exact Fisher information for censored data. The concept of hazard rate function is extended and some fundamental moment relations are established between them and the score functions. Some new moment equalities are obtained for the normal and gamma distributions.     

Exact confidence regions for scale and location parameters based on sample quartiles

In this paper we present relatively simple (ruler, paper, and pencil) nonparametric procedures for constructing joint confidence regions for (i) the median and the inner quartile range for the symmetric one-sample problem and (ii) the shift and ratio of scale parameters for the two-sample case. Both procedures are functions of the sample quartiles and have exact confidence levels when the populations are continuous. The one-sample case requires symmetry of first and third quartiles about the median.

The confidence regions we propose are always convex, nested for decreasing confidence levels and are compact for reasonably large sample sizes. Both exact small sample and approximate large sample distributions are given.     

Variance component estimation in multiple regression models having a nested error structure

Multiple regression when regressors are measured on two different sized experimental units involves a nested error structure. This nested error structure consists of two variance components. Sufficient conditions are presented under which UMVU estimators of these variance components exist. When these conditions are not met, two alternative estimators for the two variance components are considered and compared when possible.

This paper considers multiple regression models when regressor variables are associated with different sized experimental units resulting in a nested error structure. Nested error structures occur because of restrictions placed on randomizations. This results in experiments similar to splitplot type experiments which involove two different sizes of experimental units. Data resulting from these type of experiments consists of measurements made on larger sized experimental units as well as measurements made on smaller sized experimental units. Split-plot type experiments occur when certain combinations of the treatment factors are randomly assigned to larger sized experimental units after which these units are split or divided and other combinations of the treatment factors are randomly assigned to the split units.     

With the best of intentions: Singapore volunteers and donors and the tsunami reconstruction

Charities and humanitarian organisations in Singapore that were involved in post-Tsunami relief and reconstruction work are repeatedly queried by volunteers, donors, and journalists. The world has pledged and donated billions to Tsunami relief and reconstruction, and I have personally given money to Singapore charities. Why do I still read about needy people in temporary shelters? Where has the money gone? Another question reflects similar sentiments of puzzlement, even frustration: I am moved by the devastation and would like to help by volunteering my time and expertise. I have signed up to volunteer with several humanitarian organisations. I am highly qualified and could help people. Why am I still not sent overseas? These are good questions that deserve solid answers. This article discusses the latest thinking in relief circles as to which volunteers are best sent overseas and how donations can most responsibly be used for effective and sustainable reconstruction. The authors draw on Singapore's experience with providing relief in the aftermath of the Tsunami along with the Dr. Tan Tay Keong's observations of relief and rehabilitation efforts in the Tsunami-affected areas of Sri Lanka and Indonesia. The article finds that volunteering requires much more than just a giving heart. Also, donations may be used most effectively by thinking long term rather than just short term. Further, donations are best administered in a way coordinated with local relief plans and in consultation with local people. The article closes by considering lessons for volunteers, donors, and humanitarian relief agencies from the Tsunami's aftermath.     

Reviews

Abstract

Jill M. Church reviews Exploring Discovery: The Front Door to Your Library's Licensed and Digitized Content; Scott Johnson reviews Now You're a Manager: Quick and Practical Strategies for New Mid-level Managers in Academic Libraries.