Data Envelopment Analysis for Composite Indicators: A Multiple Layer Model

The development of a composite indicator (CI) over a set of individual indicators is worthwhile in case the methodological aggregation process is sound and the results are clear. It can then be used as a powerful tool for performance evaluation, benchmarking, and decision making. In this respect, data envelopment analysis (DEA), as a self appraisal technique, has recently received considerable attention in the construction of CIs for policy analysis and public communication. However, due to the ever increasing complexity of numerous performance evaluation problems, more and more potential indicators might be developed to represent an evaluation activity in a more comprehensive way. These indicators might also belong to different categories and further be linked to one another constituting a multilayer hierarchical structure. Simply treating all the indicators to be in the same layer as is the case in the basic DEA model thereby ignores the information on their hierarchical structure, and further leads up to weak discriminating power and unrealistic weight allocations. To overcome this limitation, a multiple layer DEA-based CI model is developed in this study to embody a hierarchical structure of indicators in the DEA framework, and both its primal and dual form are realized. The proposed model is illustrated by constructing a composite road safety performance index for a set of European countries.     

EXPOSING ADOPTION MYTHS: ACCESS TO INFORMATION ABOUT ORIGINS IN VICTORIA

Victoria was the first State to allow adopted persons unqualified access to their original identity and to provide natural parents and others with conditional access to identifying information. This article examines findings from 1985 to 1988 to examine four assumptions. These are that: few adoptees would want to know about or meet their natural parents and so few would register with the Adoption Information Service; few adoptees would welcome an approach from or on behalf of their natural parents and most would reject such an approach; few relinquishing parents would wish to meet or know about their relinquished child and consequently most would reject an approach by or on behalf of their child; and adoptive parents would not wish to acknowledge the adoptive status of their children by providing information to the natural parent facilitating a meeting between the natural parent and the child. The data suggest that the above assumptions do not reflect the experience or behaviour of adopted persons, natural parents or adoptive parents.     

Towards a behavioral theory of systemic hypothesis-testing and the error of the third kind

Scientific ideas neither arise nor develop in a vacuum. They are always nutured against a background of prior, partially conflicting ideas. Systemic hypothesistesting is the problem of testing scientific hypotheses relative to various systems of background knowledge. This paper shows how the problem of systemic hypothesis-testing (Sys HT) can be systematically expressed as a constrained maximimization problem. It is also shown how the error of the third kind (E _III) is fundamental to the theory of Sys HT.The error of the third kind is defined as the probability of having solved the ‘wrong’ problem when one should have solved the ‘right’ problem. This paper shows howE _III can be given both a systematic as well as a systemic treatment. Sys HT gives rise to a whole host of new decision problems, puzzles, and paradoxes.     

Statistical Tools on the World Wide Web

The World Wide Web (WWW) represents a powerful tool for furthering the development and practice of statistics. The GASP (Globally Accessible Statistical Procedures) WWW site has been set up as a primary listing of statistical procedures which can be used over the WWW. This article highlights several possible approaches for making a procedure WWW accessible. These approaches effectively solve many of the problems typically encountered when using a new statistical procedure. Applying the methods discussed, any statistical technique can be made available to anyone with a forms- or Java-capable WWW browser. Procedures can be delivered in a virtually platform-independent manner with only minimal requirements on a user's hardware or software.     

Stability of Approximations of Average Run Length of Risk-Adjusted CUSUM Schemes Using the Markov Approach: Comparing Two Methods of Calculating Transition Probabilities

Risk-adjusted CUSUM schemes are designed to monitor the number of adverse outcomes following a medical procedure. An approximation of the average run length (ARL), which is the usual performance measure for a risk-adjusted CUSUM, may be found using its Markov property. We compare two methods of computing transition probability matrices where the risk model classifies patient populations into discrete, finite levels of risk. For the first method, a process of scaling and rounding off concentrates probability in the center of the Markov states, which are non overlapping sub-intervals of the CUSUM decision interval, and, for the second, a smoothing process spreads probability uniformly across the Markov states. Examples of risk-adjusted CUSUM schemes are used to show, if rounding is used to calculate transition probabilities, the values of ARLs estimated using the Markov property vary erratically as the number of Markov states vary and, on occasion, fail to converge for mesh sizes up to 3,000. On the other hand, if smoothing is used, the approximate ARL values remain stable as the number of Markov states vary. The smoothing technique gave good estimates of the ARL where there were less than 1,000 Markov states.     

Scientization Under Pressure—The Problematic Role of Expert Bodies During the Handling of the COVID-19 Pandemic

Public Organization Review - This article focuses on the role of experts in the Norwegian decision-making process in central government during the crisis management of the COVID-19 pandemic. It is...     

On the Measurement of Corporatism

ABSTRACT: It is noticed that in both theoretical and empirical analyses concerning the relationship between corporatism and economic performance, the role of government is kept rather implicit. This article aims to incorporate government into the analysis in a more explicit way by analyzing whether government policies differ between corporatist and non-corporatist countries. The empirical results of the research reveal that corporatist countries are not only characterized by centralization or coordination of wage bargaining, but also by relatively high government expenditure, tax revenues, a higher ratio of active to passive labour market policies and also higher average unemployment benefits.     

Daisee: Adaptive importance sampling by balancing exploration and exploitation

We study adaptive importance sampling (AIS) as an online learning problem and argue for the importance of the trade-off between exploration and exploitation in this adaptation. Borrowing ideas from the online learning literature, we propose Daisee, a partition-based AIS algorithm. We further introduce a notion of regret for AIS and show that Daisee has $𝒪(\sqrt{T}{(\log T)}^{\frac{3}{4}})$ cumulative pseudo-regret, where $T$ is the number of iterations. We then extend Daisee to adaptively learn a hierarchical partitioning of the sample space for more efficient sampling and confirm the performance of both algorithms empirically.