DISTRIBUTION-FREE TESTS BASED ON SUB-SAMPLE EXTREMA FOR TESTING AGAINST POSITIVE DEPENDENCE

A class of distribution-free tests based on U-statistics has been proposed for testing the null hypothesis of independence against positive quadrant dependence. The tests are based on U-statistics and the Kendall's-tau test belongs to this class.     

JACKKNIFE TESTS FOR HETEROSCEDASTICITY IN THE GENERAL LINEAR MODEL

In this paper, tests based on the Jackknife technique are proposed to test for heteroscedasticity in the linear regression model when the errors are non-normal. These are the Jackknifed Goldfeld-Quandt (GQ), and jack-knife related variations of White (H), Lagrange multiplier (LM), Glejser (GL) and Bickel (B) tests. The power of the proposed tests is compared with that of GQ, H, LM, GL and B tests; and the robustness to the error distribution is analyzed under several heteroscedastic assumptions. The GQ test is by far the best test if the error distribution is close to normal, however, GQ test is not robust against non-normal errors. By applying the jackknife technique to the regression a more robust statistic (GQJRG) is produced but the cost is a loss in power. The GQJRG statistic generally is not M powerful as the Bickel (BlOLS) and Glejser (GLlOLS) statistics.     

A comparative study on detection of influential observations in linear regression

A large number of statistics are used in the literature to detect outliers and influential observations in the linear regression model. In this paper comparison studies have been made for determining a statistic which performs better than the other. This includes: (i) a detailed simulation study, and (ii) analyses of several data sets studied by different authors. Different choices of the design matrix of regression model are considered. Design A studies the performance of the various statistics for detecting the scale shift type outliers, and designs B and C provide information on the performance of the statistics for identifying the influential observations. We have used cutoff points using the exact distributions and Bonferroni's inequality for each statistic. The results show that the studentized residual which is used for detection of mean shift outliers is appropriate for detection of scale shift outliers also, and the Welsch's statistic and the Cook's distance are appropriate for detection of influential observations.     

Severe self-neglect: an epidemiological and historical perspective

Older adults with severe self-neglect have multiple deficits in various social, functional and physical domains, and often live insqualor. These individuals often present with poor personal hygiene, domestic squalor and hoarding which results in a threat to their own health as well as personal and public safety. Severe self-neglect occurs along a continuum with older adults often having cognitive and affective disorders compared with younger individuals presenting with psychiatric illnesses. In cases of severe self-neglect with hoarding, evidence has shown this behavior occurs in diverse social strata and not among the wealthy and professionals alone as believed earlier. Due to the multiple conditions associated with severe self-neglect, this population will require an interdisciplinary, multidimensional approach to reduce morbidity and mortality rates including nursing home placement. Research on this group has been limited and is rarely described in the medical and nursing literature. Future research is needed to provide practitioners with effective screening instruments and interventions on cases of severe self-neglect.     

A two-stage procedure for selecting the best of k exponential distributions

The problem of selecting the best of k exponential distributions with different guarantee times and the same unknown variance is considered. A two-stage procedure, similar to the one considered by Bechhofer, Dunnett and Sobel (1954), is given. Some specific guidelines for selecting the first-stage sample size are also given.     

On detection of outliers in symmetric normal models

Extensions of recent results for detection of mean slippage type outliers from i.i.d. multivariate normal and elliptically symmetric distributions are made to symmetric case, that is, when the observations are equicorrelated. The main tool used is Wijsman's (1967) representation theorem. The results obtained can be viewed as a robustness property of the use of Mardia's multivariate kurtosis as a locally optimal test statistic to detect outliers against equicorrelated distributions.     

An algorithm for the minimum sum of weighted absolute errors regression

We propose an algorithm to estimate the unknown constants in a multiple linear regression model under the minimum sum of weighted absolute errors (MSWAE). The proposed algorithm, a generalization of an earlier algorithm, is compared to a bounded variable algorithm. Some somputational experience is reported.     

Adolescents in residential treatment: The prevalence of ACEs,substance use and justice involvement

Abstract

Although associations between developmental trauma, juvenile justice involvement and youth substance use have been previously reported in the published literature, the interconnection among these three factors has not been adequately studied. This article describes the interconnection among these three factors and calls for greater attention to the role of adverse childhood experiences (ACEs) in the diagnosis and treatment of youth who present with histories of substance use and/or offending behavior. Pilot data are presented that show high rates of self-reported trauma exposure, substance use history, justice involvement, and mental health problems in a sample of adolescents in residential treatment. The data point to the need for residential treatment centers to consider trauma histories in developing treatment plans for youth with current and prior substance use and offending behavior.