Asymptotics for Statistical Treatment Rules

This paper develops asymptotic optimality theory for statistical treatment rules in smooth parametric and semiparametric models. Manski (2000, 2002, 2004) and Dehejia (2005) have argued that the problem of choosing treatments to maximize social welfare is distinct from the point estimation and hypothesis testing problems usually considered in the treatment effects literature, and advocate formal analysis of decision procedures that map empirical data into treatment choices. We develop large‐sample approximations to statistical treatment assignment problems using the limits of experiments framework. We then consider some different loss functions and derive treatment assignment rules that are asymptotically optimal under average and minmax risk criteria.     

Excluded losses and the demand for insurance

The demand for insurance is examined when the insured asset can incur losses that are excluded from insurance coverage. These losses are negatively correlated with covered losses and hence cannot be treated as background risk. Excluded losses have strikingly different effects on the demand for insurance than does background risk and lead to a modification of many standard insurance demand results. A number of new theorems concerning the effects of excluded losses are also presented. Risk-averse and prudent decision makers reduce their demand for insurance when excluded losses increase in size or riskiness. Excluded losses are a possible explanation for why many decision makers fail to take up insurance when it is offered.     

Surprise,surprise: the New Zealand general election of 2017

ABSTRACT

This article analyses the 2017 general election in New Zealand and draws out some of its implications. The final outcome was a surprise, and its implications momentous. Events strengthened Labour under a new leader and weakened the Green and New Zealand First parties, making it possible for Labour to become a viable coalition formateur. Comparing campaign public polling results, at least some of the polls were ‘wrong’, but all added to the excitement and uncertainty. Coupled with a new Labour Party leader, that uncertainty and excitement had mobilising effects that affected the young, but not the old. Finally, this article analyses the changes to the party system, nation-wide, and in the Māori electorates, assessing the degree of two-party dominance, and draws out implications for debates about the threshold for representation.     

The role of the state in structural transition and economic crisis

As academicians, we emphasize to our students to keep up with class, and always read ahead. When I took over as President-elect of the WSSA in April of 2008, while busy communicating with the Section Coordinators and plan the 2009 Annual Conference (Reno, Nevada), I would occasionally pause and ponder what I would present for my President Address (reading ahead). What do I know that would be of interest to the general membership of the WSSA? I naturally thought of China and its economic reform, as half of my research output over the last decade has been in this area. Then, the subprime crisis came to be, and I wondered about how the economy would cope. I can claim some expertise to the former, but of no special authority regarding the latter. However, I do have some thoughts regarding the role of the State during such economic upheavals, but in a more conceptual manner than a policy solution.     

Patterns of change in religious service attendance across the life course: Evidence from a 34-year longitudinal study

Although a number of studies have uncovered evidence of age differences in religious involvement across the life course, there has been a lack of long-term longitudinal data to test the extent to which these differences are due to changes within individuals over time. This study tracks trajectories of change in religious service attendance using data collected longitudinally over the course of up to 34 years, between 1971 and 2005, and in ages ranging from 15 to 102. Piecewise growth curve modeling was used to examine changes in the patterns of age-related change in three distinct developmental periods: the transition from adolescence to young adulthood, middle adulthood, and older adulthood. Attendance showed an average pattern of quadratic decline in adolescence, stability in middle adulthood, and a quadratic pattern of more rapid increase followed by decrease over the course of older adulthood. These results suggest that developmental factors play a role in changing patterns of religious participation across the adult life course, and may account for some of the apparent differences between age groups.     

Normalized measures of concavity and Ross’s strongly more risk averse order

The Arrow-Pratt (A-P) definitions of absolute and relative risk aversion dominate the discussion of risk aversion and defining “more risk averse”. Ross (Econometrica 49:621–663, 1981) notes, however, that being A-P more risk averse is not sufficient for addressing many important comparative static questions. Consequently he introduces “a new and stronger measure for comparing two agents’ attitudes towards risk…”. Ross does not provide a corresponding measure of risk aversion. This paper uses a normalized measure of concavity to characterize the Ross definition of strongly more risk averse on bounded intervals. Other properties and uses of these normalized measures of concavity are also presented.     

Impossibility Results for Nondifferentiable Functionals

We examine challenges to estimation and inference when the objects of interest are nondifferentiable functionals of the underlying data distribution. This situation arises in a number of applications of bounds analysis and moment inequality models, and in recent work on estimating optimal dynamic treatment regimes. Drawing on earlier work relating differentiability to the existence of unbiased and regular estimators, we show that if the target object is not differentiable in the parameters of the data distribution, there exist no estimator sequences that are locally asymptotically unbiased or α‐quantile unbiased. This places strong limits on estimators, bias correction methods, and inference procedures, and provides motivation for considering other criteria for evaluating estimators and inference procedures, such as local asymptotic minimaxity and one‐sided quantile unbiasedness.