Parameter estimation in high dimensional Gaussian distributions

In order to compute the log-likelihood for high dimensional Gaussian models, it is necessary to compute the determinant of the large, sparse, symmetric positive definite precision matrix. Traditional methods for evaluating the log-likelihood, which are typically based on Cholesky factorisations, are not feasible for very large models due to the massive memory requirements. We present a novel approach for evaluating such likelihoods that only requires the computation of matrix-vector products. In this approach we utilise matrix functions, Krylov subspaces, and probing vectors to construct an iterative numerical method for computing the log-likelihood.     

Why do American Indians vote Democratic?

We examine the relationship between a relatively large population of American Indians in eastern Oklahoma and the slow pace of that region's partisan realignment. With a 2012 exit poll, we empirically examine whether American Indians’ cultural ties and their socio-economic characteristics affected their vote choices. Because of their greater poverty, American Indians are more economically progressive and vote more Democratic than their white counterparts.     

Joint measurement of risk aversion,prudence, and temperance

Risk aversion—but also the higher-order risk preferences of prudence and temperance—are fundamental concepts in the study of economic decision making. We propose a method to jointly measure the intensity of risk aversion, prudence, and temperance. Our theoretical approach is to define risk compensations of different orders, and in an experiment we elicit these compensations with a price list technique. We find evidence for risk aversion, prudence, and temperance. These traits correlate within subjects. The compensations elicited for prudence are significantly larger than those for risk aversion and temperance. In contrast to commonly used utility functions, prospect theory can predict this behavioral pattern. In our experiment, risk-averse, risk-loving, and risk-neutral subjects are prudent. This supports a recent theoretical observation that prudence may be a more universal trait than previously realized.     

Identifying the Material and Non‐Material Ingredients of the Good Life

Criticism of the neo‐liberal approach to generating economic growth and rising prosperity rests on the view that it has failed to deliver many of the things that people most value. Developments in the measurement of quality of life, and in the poverty literature, have moved away from relying solely on economic variables such as income. The deprivation approach provides an insight into the items that people value because it asks a random sample of the population whether or not a list of items are regarded as essential – things that no‐one should have to go without. Drawing on results from a recent deprivation survey, this paper examines the strength of Australian opinion about whether material and non‐material items are essential. The results confirm that there is widespread support for many non‐material items being essential and that lack of ownership of these items is often widespread, with adverse affects on wellbeing.     

Low Fertility and Long-Run Growth in an Economy with a Large Public Sector

Recently it has been suggested that low fertility countries may be caught in a trap that is hard to get out of. One important mechanism in such a trap would be social interaction and its effect on the ideal family size. Such social interaction mechanisms are hard to capture in formal models, therefore we use an agent-based simulation model to investigate the issue. In our experimental setup a stable growth and population path is calibrated to Swedish data using the Swedish social policy setup. The model is provoked into a fertility trap by increasing relative child costs linked to positive growth. Even rather large increases in child benefits are then insufficient to get out of the trap. However, the small number of children temporarily enables the economy to grow faster for several decades. Removing the adaptation of social norms turns out to disarm the trap.     

Improved adaptive group testing algorithms with applications to multiple access channels and dead sensor diagnosis

We study group-testing algorithms for resolving broadcast conflicts on a multiple access channel (MAC) and for identifying the dead sensors in a mobile ad hoc wireless network. In group-testing algorithms, we are asked to identify all the defective items in a set of items when we can test arbitrary subsets of items. In the standard group-testing problem, the result of a test is binary—the tested subset either contains defective items or not. In the more generalized versions we study in this paper, the result of each test is non-binary. For example, it may indicate whether the number of defective items contained in the tested subset is zero, one, or at least two. We give adaptive algorithms that are provably more efficient than previous group testing algorithms. We also show how our algorithms can be applied to solve conflict resolution on a MAC and dead sensor diagnosis. Dead sensor diagnosis poses an interesting challenge compared to MAC resolution, because dead sensors are not locally detectable, nor are they themselves active participants. A preliminary version of this paper was presented at SPAA 2006.     

Structuring mental health within senior services: a case study of an outpatient mental health clinic serving latinos

The article presents a case study of an outpatient mental health clinic serving Latino older adults. The study explored staff perceptions on the clinic development and the context within an immigrant multicultural community. The study used in-depth, semistructured interviews with support staff, clinical social workers, and administrators. Interviews were analyzed using a thematic content analysis. The mental health clinic formation was perceived by staff as an ecological process stemming from the needs of the Latino immigrant senior clients. A close knit and interdependent culture allowed the clinic to adjust to diversity and changing cultural contexts.     

The influence of socio-economic characteristics on the political attitudes of American Indians

We examine the political attitudes of American Indians in eastern Oklahoma where they make up almost 20% of the population. We argue that American Indians in the region play the same role that other minorities in the southern states—notably blacks and Hispanics—do in resisting the region's realignment from the Democratic to the Republican Party. American Indians in the region are populists in that they are economically liberal and religiously conservative. The results of our analysis suggest that the socio-economic characteristics of American Indians play a greater role in explaining American Indians’ political attitudes than their ethnic identification.     

Belief elicitation in the presence of na?ve respondents: An experimental study

It is often of interest to elicit beliefs from populations that may include na?ve participants. Unfortunately, elicitation mechanisms are typically assessed by assuming optimal responses to incentives. Using laboratory experiments with a population that potentially includes na?ve participants, we compare the performance of two elicitation mechanisms proposed by Karni (Econometrica 77(2):603-606, 2009). These mechanisms, denoted as “declarative” and “clock,” are valuable because their incentive compatibility does not require strong assumptions such as risk neutrality or expected utility maximization. We show that, theoretically and empirically, with a sufficient fraction of na?ve participants, the clock mechanism elicits beliefs more accurately than the declarative. The source of this accuracy advantage is twofold: the clock censors na?ve responses, and participants are more likely to employ dominant strategies under the clock. Our findings hold practical value to anyone interested in eliciting beliefs from representative populations, a goal of increasing importance when conducting large-scale surveys or field experiments.