ORDER SELECTION IN ARMA MODELS USING THE FOCUSED INFORMATION CRITERION

This paper develops a new approach for order selection in autoregressive moving average models using the focused information criterion. This criterion minimizes the asymptotic mean squared error of the estimator of a parameter of interest. Simulation studies indicate that the suggested criterion is quite effective and comparable to the Akaike information criterion, the corrected Akaike information criterion and the Bayesian information criterion in autoregressive moving average order selection. The use of the focused information criterion for the simultaneous selection of regression variables and order of the error process in a linear regression model with autoregressive moving average errors is also considered.     

Marriages of bisexual men

This study examined the marriages of 26 couples in which the husband was bisexual. The subjects were a non-clinical sample married for at least two years and intending to continue their marriages. The sample was, overall, highly educated and earned concomitantly high incomes. Subjects were administered the Klein Sexual Orientation Grid and a research questionnaire to determine successful or problematic aspects of their marriages. Subjects were, for the most part, satisfied with the quality of their marriages, sexually active within the marriages, and open about the husband's homosexual behavior. A high-level of sexual activity within the marriage, open and direct communication, a valued friendship, previous counseling or psychotherapy, cognitive flexibility, and financial independence contributed to the success of these marriages. The husbands reported a great deal of ambiguity about their homosexual behavior, and the couples reported intense conflict dealing with their open marriage styles.     

Long- and short-term affective states in happiness: Model,paradigm and experimental evidence

The present paper evaluates relevant findings on long- and short-term affective states in subjective well-being and argues for a componential model that combines the two into the more general concept of happiness. Two age parameters, one for long-term and one for short-term affect are added to the simple model to account for age changes in happiness. Measures of long- and short-term affect are presented, as is a paradigm for separating the components. Support for the simple model is provided by an experimental investigation with 64 college students who were exposed to positive and/or negative mood induction. As expected, the induction had substantially greater effect on short- than on long-term affect measures, particularly the negative induction procedure. These findings are consistent with predictions. However, multiple age cohorts will have to be assessed within a longitudinal framework to obtain values for the age parameters.     

CORRECTING FOR MEASUREMENT ERROR IN AN EXPOSURE-RESPONSE RELATIONSHIP BASED ON DICHOTOMISING A CONTINUOUS DEPENDENT VARIABLE

Random error in a continuous outcome variable does not affect its regression on a predictor. However, when a continuous outcome variable is dichotomised, random measurement error results in a flatter exposure-response relationship with a higher intercept. Although this consequence is similar to the effect of misclassification in a binary outcome variable, it cannot be corrected using techniques appropriate for binary data. Conditional distributions of the measurements of the continuous outcome variable can be corrected if the reliability coefficient of the measurements can be estimated. An unbiased estimate of the exposure-response relationship is then easily calculated. This procedure is demonstrated using data on the relationship between smoking and the development of airway obstruction.     

Rejoinder to 'Ahmed, M.S. (1998). A note on regression-type estimators using multiple auxiliary information.'

In the estimators t ₃ , t ₄ , t ₅ of Mukerjee, Rao & Vijayan (1987), b _{y x} and b _{y z} are partial regression coefficients of y on x and z , respectively, based on the smaller sample. With the above interpretation of b _{y x} and b _{y z} in t ₃ , t ₄ , t ₅ , all the calculations in Mukerjee at al. (1987) are correct. In this connection, we also wish to make it explicit that b _{x z} in t ₅ is an ordinary and not a partial regression coefficient. The 'corrected' MSEs of t ₃ , t ₄ , t ₅ , as given in Ahmed (1998 Section 3) are computed assuming that our b _{y x} and b _{y z} are ordinary and not partial regression coefficients. Indeed, we had no intention of giving estimators using the corresponding ordinary regression coefficients which would lead to estimators inferior to those given by Kiregyera (1984). We accept responsibility for any notational confusion created by us and express regret to readers who have been confused by our notation. Finally, in consideration of the above, it may be noted that Tripathi & Ahmed's (1995) estimator t ₀ , quoted also in Ahmed (1998), is no better than t ₅ of Mukerjee at al. (1987).     

Mental health, social relations, and social selection: a longitudinal analysis.

A wide body of literature documents the effect of social networks and social supports on mental health. Fewer studies, however, have examined the reciprocal effect of mental health on social relationships. This problem is examined using data from a national panel survey of adults aged 20-64. For the sample as a whole, support was found for a social selection process, since psychological distress predicted decreases in primary, but not secondary, social relationships. The extent of primary relationships also were found to be associated with subsequent distress, providing evidence that the relationship between mental health and social environment may be transactional. When examined separately by gender, males but not females were found to be vulnerable to the process of social selection, supporting the hypothesis that the expression of distress is less role-appropriate for men and therefore more likely to invite social sanctions. Social causation effects also were observed only among males.     

Summarizing the Effect of a Wide Array of Amenity Measures into Simple Components

A significant issue existing within the rural economic development literature revolves around the difficulty with sorting out the controversy of the effects of amenity activities on rural economic growth. This problem is due to the different ways amenity attributes are linked to regional economic performance. Numerous researchers utilize principal component analysis to compress groups of variables that describe attributes of natural-based amenity and quality of life into scalar measures. While principal components are good at reducing a collection of variables into single measures, they often lack interpretability because they define some abstract scores which are often not meaningful or not well interpretable in practice. We apply the simple component analysis suggested by Rousson and Gasser (2004, Applied Statistics 53, 539–555) to summarize the information in groups of variables into a limited number of simple components and improve interpretability at a modest loss of optimality. Simple components allow us to identify and interpret the effect of attributes that most influence regional economic performance so as to gain better insight into policies to preserve and advance those attributes. The same methodology is appropriate for any social science discipline when there is a need to replace a larger number of multiple indicator measurements with a smaller set.