Higher Order Properties of Gmm and Generalized Empirical Likelihood Estimators

In an effort to improve the small sample properties of generalized method of moments (GMM) estimators, a number of alternative estimators have been suggested. These include empirical likelihood (EL), continuous updating, and exponential tilting estimators. We show that these estimators share a common structure, being members of a class of generalized empirical likelihood (GEL) estimators. We use this structure to compare their higher order asymptotic properties. We find that GEL has no asymptotic bias due to correlation of the moment functions with their Jacobian, eliminating an important source of bias for GMM in models with endogeneity. We also find that EL has no asymptotic bias from estimating the optimal weight matrix, eliminating a further important source of bias for GMM in panel data models. We give bias corrected GMM and GEL estimators. We also show that bias corrected EL inherits the higher order property of maximum likelihood, that it is higher order asymptotically efficient relative to the other bias corrected estimators.     

INDUSTRY STUDIES AND AMERICAN INDUSTRIAL PERFORMANCE

A nationwide research program on the performance of American industry, sponsored by the Alfred P. Sloan Foundation, now involves scores of faculty researchers and more than 100 graduate students at several American universities. Ten research centers have been established, and each one is dedicated to the study of a particular industry. An eleventh center is carrying out research on problems that cut across these industries. It is expected that the Sloan Industry Studies program will make useful contributions to the performance of the industries themselves, to policymaking at all levels of government, and to the development of interdisciplinary research communities within the universities.     

Extended-family resources and racial inequality in the transition to homeownership

We use longitudinal data from the Panel Study of Income Dynamics to examine how access to financial resources in the extended family affects the accumulation of wealth among non-owners and how these resources subsequently affect transitioning into homeownership. Our findings show that economic conditions of the extended family have substantial effects on non-owners’ wealth accumulation and likelihood of becoming homeowners, even after adjusting for individual sociodemographic and economic characteristics. We find significant effects of extended-family wealth for both black and white households, but effects of extended-family income insufficiency for blacks only. Consequently, limited access to wealth and greater level of poverty in the extended family hamper blacks’ transition to homeownership. Our results show that the level of extended-family wealth necessary for black householders to equalize their likelihood of becoming homeowners with whites is very high. In fact, our findings indicate that white householders embedded in extended families with no net wealth are just as likely to make the transition to ownership as are black householders with affluent extended families. These findings support arguments related to the importance of extended-family resources in processes of residential attainment but also point to important racial differences in not only levels but also consequences of these family resources.     

THE IMPACT OF HUMAN ERROR ON DELIVERING SERVICE QUALITY

This paper opens a new avenue for investigation of quality issues in services. We take the viewpoint that a substantial portion of service failures is the result of human error in the delivery process. Drawing upon the Generic Error Modeling System (gems) from the cognitive science literature, we develop a framework for understanding the role of human error in service failures. An empirical investigation assesses the applicability of this framework to services, identifies which error mechanisms are important sources of service failure, and clarifies how the different roles of customers and providers affect the errors made by each.     

A COMPARISON OF PRODUCTION SCHEDULING POLICIES ON COSTS,SERVICE LEVEL,AND SCHEDULE CHANGES

We consider a single product, single level, stochastic master production scheduling (Mps ) model where decisions are made under rolling planning horizons. Outcomes of interest are cost, service level, and schedule stability. The subject of this research is the Mps control system: the method used in determining the amount of stock planned for production in each time period. Typically, Mps control systems utilize a single buffer stock. Here, two Mps dual-buffer stock systems are developed and tested by simulation. We extend the data envelopment analysis (dea ) methodology to aid in the evaluation of the simulation results, where Dea serves to increase the scope of the experimental design. Results indicate that the dual-buffer control systems outperform existing policies.     

The Auditor's Going-Concern Disclosure as a Self-Fulfilling Prophecy: A Discrete-Time Survival Analysis*

The question of how an auditor's going-concern disclosure affects a client's future operations has long troubled the auditing profession. In an attempt to provide further understanding of this issue, we introduce Discrete-Time Survival Analysis (DTSA) to examine the aftermath of 231 first-time going-concern disclosures on clients' subsequent continuance. DTSA represents a significant refinement over traditional ordinary least squares (OLS) and logistic (LOGIT) regression in that it provides not only a probability estimate, but also an estimate of the timing of the event occurrence. The addition of this extra dimension (event timing) aids decision makers by providing more complete information about event probabilities. Consistent with the “self-fulfilling prophecy effect,” the risk profiles developed from DTSA indicate that the first year subsequent to the initial going-concern disclosure was the most dangerous in terms of risk of bankruptcy. However, after the first year, the incidence of bankruptcy decreases significantly. Thus, DTSA is able to provide a richer perspective on this perplexing issue than previously considered.     

Agriculture,Pesticide Use,and Economic Development: A Global Examination (1990–2014)

Modern agricultural production typically requires large quantities of chemical pesticides, a potential source of both environmental and human harm. Previous social science research has suggested that environmental problems such as those associated with pesticide use may begin to decline at higher levels of economic development. Using fixed effects models, we examine whether this possible relationship holds within nations and over time. This study draws on data from the World Bank as well as pesticide use data from the Food and Agriculture Organization of the United Nations to examine the relationship between pesticide use and economic development within nations from 1990 to 2014. The findings are considered from theoretical perspectives in environmental sociology on the drivers of environmental impacts: the treadmill of production theory, structural human ecology, ecological modernization theory, and the environmental Kuznets curve hypothesis. The results of this study show a positive relationship between economic development and pesticide consumption over time, with no decline in use at higher levels of economic development. Thus, they generally support the claims made by treadmill of production and structural human ecology.     

Before,Now, and After: Assessing Hurricane Katrina Relief

We assess governmental and non-governmental responses to disasters using primary data of Hurricane Katrina survivors along the Mississippi Gulf Coast. Non-governmental sources include nonprofit relief groups, faith-based organizations, and survivors’ self-identified social networks. We assess the impact of these governmental and non-governmental relief efforts on survivors’ economic, psychological, physical, and social effects from the disaster. Our results show that social isolation significantly increases perceptions of disaster disturbance and decreases perceived rates of disaster relief. Additionally, survivors perceive that social networks provide greater sources of psychological, financial and social disaster relief than government sources. However, survivors’ social networks decay sharply in the immediate aftermath of a disaster, and they do not appear to fully recover a year from the disaster. These social networks themselves are not fully resilient to a disaster.

The Rheumatoid Arthritis Drug Development Model: a case study in Bayesian clinical trial simulation

The development of a new drug is a major undertaking and it is important to consider carefully the key decisions in the development process. Decisions are made in the presence of uncertainty and outcomes such as the probability of successful drug registration depend on the clinical development programmme. The Rheumatoid Arthritis Drug Development Model was developed to support key decisions for drugs in development for the treatment of rheumatoid arthritis. It is configured to simulate Phase 2b and 3 trials based on the efficacy of new drugs at the end of Phase 2a, evidence about the efficacy of existing treatments, and expert opinion regarding key safety criteria. The model evaluates the performance of different development programmes with respect to the duration of disease of the target population, Phase 2b and 3 sample sizes, the dose(s) of the experimental treatment, the choice of comparator, the duration of the Phase 2b clinical trial, the primary efficacy outcome and decision criteria for successfully passing Phases 2b and 3. It uses Bayesian clinical trial simulation to calculate the probability of successful drug registration based on the uncertainty about parameters of interest, thereby providing a more realistic assessment of the likely outcomes of individual trials and sequences of trials for the purpose of decision making. In this case study, the results show that, depending on the trial design, the new treatment has assurances of successful drug registration in the range 0.044–0.142 for an ACR20 outcome and 0.057–0.213 for an ACR50 outcome. Copyright © 2009 John Wiley & Sons, Ltd.     

Analytic bounds on causal risk differences in directed acyclic graphs involving three observed binary variables

We apply a linear programming approach which uses the causal risk difference (R_DC)$(R{D}_C)$ as the objective function and provides minimum and maximum values that R_DC$R{D}_C$ can achieve under any set of linear constraints on the potential response type distribution. We consider two scenarios involving binary exposure X, covariate Z and outcome Y. In the first, Z is not affected by X, and is a potential confounder of the causal effect of X on Y. In the second, Z is affected by X and intermediate in the causal pathway between X and Y. For each scenario we consider various linear constraints corresponding to the presence or absence of arcs in the associated directed acyclic graph (DAG), monotonicity assumptions, and presence or absence of additive-scale interactions. We also estimate Z-stratum-specific bounds when Z is a potential effect measure modifier and bounds for both controlled and natural direct effects when Z is affected by X  . In the absence of any additional constraints deriving from background knowledge, the well-known bounds on R_Dc$R{D}_c$ are duplicated: -Pr(Y≠X)?R_DC?Pr(Y=X)$-\mathit{Pr}(Y\ne X)?R{D}_C?\mathit{Pr}(Y=X)$. These bounds have unit width, but can be narrowed by background knowledge-based assumptions. We provide and compare bounds and bound widths for various combinations of assumptions in the two scenarios and apply these bounds to real data from two studies.