Some empirical observations of relevance to the analysis of union wage determination

Six important empirical characteristics of the union sector need to be incorporated into future research on wage determination. These are 1) the extent of unionization, 2) the statistical correlates of unionization, 3) divergent trends in union and nonunion earnings, 4) union/nonunion wage differentials, 5) the determinants of union and nonunion wage change, and 6) wage imitation. Examination of these characteristics suggests the following about union wage determination. Union wages have advanced relative to nonunion since the mid 1950s, despite relative shrinkage of the union sector. Union wage changes show less sensitivity to business-cycle pressures than nonunion. Limited spheres of wage imitation surround certain major union negotiations. These observations can be fitted into recent analyses of wage determination under long-term employer employee relationships, and have relevance for anti-inflation policy. Research for this paper was undertaken while the author was a Senior Fellow at the Brookings Institution and was supported by a grant from the Alfred P. Sloan Foundation. Views expressed should not be attributed to the Brookings Institution, its staff or trustees.     

Answer-perturbation techniques for the protection of statistical databases

This paper presents an answer-perturbation model for the protection of statistical databases. Compared to previous approaches, this model saves both space and time since neither a copy of the original database nor an extra processing of the query set is needed. The modularity of our protection mechanism (perturbation is applied after the query processing, not during it or upon data entry, as in the other approaches) facilitates the implementation. The method presented is user-oriented and since the usability (accuracy) of the statistically altered answers can be fine-tuned, the balance between usability and protection (security) is under the control of the database administrator.     

Public Administration and Organizational Theory: Prescribing the Proper Dose

Public Organization Review - Since the founding of the field of public administration, scholars have struggled with questions related to the “publicness” of public organizations...     

Gatekeepers: How Procurement Personnel Guard Against Hybrid Accountability

Though much attention has been given to the relationship between contractor and vendor in regard to governmental accountability, little attention has been paid to the internal struggles that surface in regard to accountability in the procurement process. These informal horizontal relationships often circumvent vertical accountability controls most often thought to ensure equitable vendor selection. Similarly, though much attention has been paid to political influence in contracting, hierarchal organizational structures, and measures of performance, the contract officer-program manager relation has not been explored. Specifically, the informal tension that often forms between the program manager and contract officer has not been examined in regard to such accountability. This paper examines the relationship between the program officer and the contract officer in order to describe how hybrid public administration can affect the contracting process.     

A multivariate finite mixture latent trajectory model with application to dementia studies

Dementia patients exhibit considerable heterogeneity in individual trajectories of cognitive decline, with some patients showing rapid decline following diagnoses while others exhibiting slower decline or remaining stable for several years. Dementia studies often collect longitudinal measures of multiple neuropsychological tests aimed to measure patients’ decline across a number of cognitive domains. We propose a multivariate finite mixture latent trajectory model to identify distinct longitudinal patterns of cognitive decline simultaneously in multiple cognitive domains, each of which is measured by multiple neuropsychological tests. EM algorithm is used for parameter estimation and posterior probabilities are used to predict latent class membership. We present results of a simulation study demonstrating adequate performance of our proposed approach and apply our model to the Uniform Data Set from the National Alzheimer's Coordinating Center to identify cognitive decline patterns among dementia patients.     

A Standardized Normal-Laplace Mixture Distribution Fitted to Symmetric Implied Volatility Smiles

This article proposes to use a standardized version of the normal-Laplace mixture distribution for the modeling of tail-fatness in an asset return distribution and for the fitting of volatility smiles implied by option prices. Despite the fact that only two free parameters are used, the proposed distribution allows arbitrarily high kurtosis and uses one shape parameter to adjust the density function within three standard deviations for any specified kurtosis. For an asset price model based on this distribution, the closed-form formulas for European option prices are derived, and subsequently the volatility smiles can be easily obtained. A regression analysis is conducted to show that the kurtosis, which is commonly used as an index of tail-fatness, is unable to explain the smiles satisfactorily under the proposed model, because the additional shape parameter also significantly accounts for the deviations revealed in smiles. The effectiveness of the proposed parsimonious model is demonstrated in the practical examples where the model is fitted to the volatility smiles implied by the NASDAQ market traded foreign exchange options.     

Simultaneous Small Sample Inference for Linear Combinations of Generalized Linear Model Parameters

A method is proposed to construct simultaneous confidence intervals for multiple linear combinations of generalized linear model parameters, that uses a multivariate normal- or t-distribution together with the signed likelihood root statistic. In an application to a case study simultaneous confidence bands for logistic regression are calculated. A simulation study based on the example evaluation suggests superior performance compared to the common Wald-type approaches. The proposed methods are readily implemented in the R extension package mcprofile.