Pseudo–Modelling of Outliers for Estimation of Regression And Time Series Models

This paper documents situations where the variance inflation model for outliers has undesirable properties. The model is commonly used to accommodate outliers in a Bayesian analysis of regression and time series models. The alternative approach provided here does not suffer from these undesirable properties but gives inferences similar to those of the variance inflation model when this is appropriate. It can be used with regression, time series, and regression with correlated errors in a unified way, and adheres to the scientific principle that inference should be based on the data after obvious outliers have been discarded. Only one parameter is required for outliers; it is interpretable as the a priori willingness to remove observations from the analysis.     

Exposure Reconstruction for the TCDD-Exposed NIOSH Cohort Using a Concentration- and Age-Dependent Model of Elimination

Recent studies demonstrating a concentration dependence of elimination of 2,3,7,8-tetrachlorodibenzo-p-dioxin (TCDD) suggest that previous estimates of exposure for occupationally exposed cohorts may have underestimated actual exposure, resulting in a potential overestimate of the carcinogenic potency of TCDD in humans based on the mortality data for these cohorts. Using a database on U.S. chemical manufacturing workers potentially exposed to TCDD compiled by the National Institute for Occupational Safety and Health (NIOSH), we evaluated the impact of using a concentration- and age-dependent elimination model (CADM) (Aylward et al., 2005) on estimates of serum lipid area under the curve (AUC) for the NIOSH cohort. These data were used previously by Steenland et al. (2001) in combination with a first-order elimination model with an 8.7-year half-life to estimate cumulative serum lipid concentration (equivalent to AUC) for these workers for use in cancer dose-response assessment. Serum lipid TCDD measurements taken in 1988 for a subset of the cohort were combined with the NIOSH job exposure matrix and work histories to estimate dose rates per unit of exposure score. We evaluated the effect of choices in regression model (regression on untransformed vs. ln-transformed data and inclusion of a nonzero regression intercept) as well as the impact of choices of elimination models and parameters on estimated AUCs for the cohort. Central estimates for dose rate parameters derived from the serum-sampled subcohort were applied with the elimination models to time-specific exposure scores for the entire cohort to generate AUC estimates for all cohort members. Use of the CADM resulted in improved model fits to the serum sampling data compared to the first-order models. Dose rates varied by a factor of 50 among different combinations of elimination model, parameter sets, and regression models. Use of a CADM results in increases of up to five-fold in AUC estimates for the more highly exposed members of the cohort compared to estimates obtained using the first-order model with 8.7-year half-life. This degree of variation in the AUC estimates for this cohort would affect substantially the cancer potency estimates derived from the mortality data from this cohort. Such variability and uncertainty in the reconstructed serum lipid AUC estimates for this cohort, depending on elimination model, parameter set, and regression model, have not been described previously and are critical components in evaluating the dose-response data from the occupationally exposed populations.     

Joint impact of multiple observations on a subset of variables in multiple linear regression

In multiple linear regression analysis, each observation affects the fitted regression equation differently and has varying influences on the regression coefficients of the different variables. Chatterjee & Hadi (1988) have proposed some measures such as DSSE_ij (Impact on Residual Sum of Squares of simultaneously omitting the ith observation and the jth variable), F_j (Partial F-test for the jth variable) and F_j(i) (Partial F-test for the jth variable omitting the ith observation) to show the joint impact and the interrelationship that exists among a variable and an observation. In this paper we have proposed more extended form of those measures DSSE_IJ, F_J and F_J(I) to deal with the interrelationships that exist among the multiple observations and a subset of variables by monitoring the effects of the simultaneous omission of multiple variables and multiple observations.     

Happiness has traitlike and statelike properties: A reply to Veenhoven

Veenhoven (1994) contrasted hypotheses of whether happiness is a trait or state, concluding that it is a state variable. This article critiques the conceptual foundation of Veenhoven's paper, and examines technical deficiencies in his review of evidence. Based on previous findings and new analyses, we conclude that happiness has both traitlike and statelike properties, but that individual differences in happiness endure despite its situational reactivity, and explain greater variance than situational effects.     

Evaluation: Stakeholder-focused Criteria

Identifying and selecting suitable criteria is an important consideration in evaluating the performance of public sector organizations. Which criteria are chosen, and how they are weighted, will depend largely on the perspective being taken; different stakeholders (for example, clients, providers or commissioners) are likely to emphasize different criteria. In this paper a number of criteria are identified and an argument is developed to indicate which criteria are likely to be valued by various stakeholders. Furthermore, a number of approaches and mechanisms for evaluation are identified and the links between these, the criteria appropriate to them and the various stakeholders are made explicit. Finally, the issue of politics and the distribution of power between the stakeholders is addressed with a call for greater explicitness in using and balancing criteria (for example, as between professionals and the intended beneficiaries of services). In this way, while recognizing the inherent political nature of performance evaluation, the planning and implementation of services may be improved by a more mutually informed approach to evaluation.     

Interpersonal dependency of preferences

Standard decision theoretic models disregard the phenomenon of interpersonal dependency of preferences. In this paper it is argued that interpersonal dependency of preferences is a serious challenge for standard utility theory. First we sketch the more philosophical aspects of the problem and then, using a simple, formal model for the two-person case, we show that interpersonal dependency of preferences generally results in indeterminacy of preferences (resp. of subjective utility).     

On the derivation of a full life table from mortality data recorded in five-year age groups

Mortality data are often gathered using 5-year age groups rather than individual years of life. Furthermore, it is common practice to use a large open-ended interval (such as 85 and over) for mortality data at the older ages. These limitations of the data pose problems for the actuary or demographer who wishes to compile a full and accurate life table using individual years of life. The author devises formulae which handle these problems. He also devises methods for handling mortality during the 1st year of life and for dealing with other technical problems which arise in the compilation of the full life table from grouped data.     

Effects of migration on family structure

This paper examines the situation and problems of migration on family structure, with emphasis on family reunification. The study is based on conditions and practices in Western Europe and Mediterranean countries relating to temporary labor migration. Most migrant workers have no intention of settling permanently and return to their country within a few years. The International Labour Office estimated in 1974 that at least 1/2 the migrant workers in Western Europe live without their families. Generally, migrants send for their families only when they are employed, earning adequate wages, and have adequate housing. Some reasons why migrants live apart from their families include 1) the receiving country discourages family immigration because it does not coincide with the economic necessities of migration policy and 2) some sending countries discourage it to ensure that the migrant worker returns to his own country. The main danger arising from family separation is that it frequently leads to the break up of the family. The leading European authorities recognize as a fundamental right the freedom of a migrant worker and his family to lead a normal family life in the receiving country. The author outlines the conditions for admission for residence and employment of migrant spouses and children for the Federal Republic of Germany, France, Belgium, Switzerland, the Netherlands, Sweden, Norway, Austria, Luxembourg, and the United Kingdom. All countries require that the head be in regular employment for some time and be able to provide his family with suitable housing. Other problems concerning the arrival of migrant spouses and children include 1) acquiring employment and social information and counseling, 2) education of children, 3) obtaining vocational training and adaptation and 4) achieving entitlement to social security benefits. The effects of migration in the family context in sending countries include 1) providing activities for migrants to maintain cultural links with their countries of origin and 2) acquiring the nationality of the receiving countries. Countries should facilitate the admission to employment of migrant spouses and children, by maintaining provisions for the reuniting of families and imposing no limits on admission to residence; and 2) by overcome obstacles to admission to employment, by observing existing recommendations. In conclusion, governments should give family cohesion 1st priority, regardless of regulations.     

Enrich marital inventory: a discriminant validity and cross-validation assessment

To assess the validity and clinical utility of the marital inventory ENRICH, a discriminant validity study was conducted using a national sample of 5039 married couples. The sample was randomly split in order to form a cross-validation group. ENRICH is a multidimensional scale and two types of analysis were conducted to assess the value of these various scales. Results from discriminant analysis indicated that using either the individual scores or couples' scores, happily married couples could be discriminated from unhappily married couples with 85–95% accuracy. These results were cross-validated with a second sample. Using regression analysis, it was clearly demonstrated that background factors account for little of the variance in discriminating happy from unhappily married couples compared to their relationship dynamics, i.e., scale scores. All ENRICH scales except equalitarian roles proved significant, indicating the validity of a multidimensional inventory.