计量经济分析 = Econometric analysis

25. Multiple Imputation for Nonresponse in Surveys. By D. B. Rubin. ISBN 0 471 08705 X. Wiley, Chichester, 1987. 258 pp. £30.25.

Multiple Imputation for Nonresponse in Surveys

We discuss a method for improving causal inferences called ‘‘Coarsened Exact Matching’’ (CEM), and the new ‘‘Monotonic Imbalance Bounding’’ (MIB) class of matching methods from which CEM is derived. We summarize what is known about CEM and MIB, derive and illustrate several new desirable statistical properties of CEM, and then propose a variety of useful extensions. We show that CEM possesses a wide range of statistical properties not available in most other matching methods but is at the same time exceptionally easy to comprehend and use. We focus on the connection between theoretical properties and practical applications. We also make available easy-to-use open source software for R, Stata, and SPSS that implement all our suggestions.

/pdf/causal-inference-without-balance-checking-coarsened-exact-1xeumfa548.pdf

Causal Inference without Balance Checking: Coarsened Exact Matching

We explore the effect of dimensionality on the nearest neighbor problem. We show that under a broad set of conditions (much broader than independent and identically distributed dimensions), as dimensionality increases, the distance to the nearest data point approaches the distance to the farthest data point. To provide a practical perspective, we present empirical results on both real and synthetic data sets that demonstrate that this effect can occur for as few as 10-15 dimensions. These results should not be interpreted to mean that high-dimensional indexing is never meaningful; we illustrate this point by identifying some high-dimensional workloads for which this effect does not occur. However, our results do emphasize that the methodology used almost universally in the database literature to evaluate high-dimensional indexing techniques is flawed, and should be modified. In particular, most such techniques proposed in the literature are not evaluated versus simple linear scan, and are evaluated over workloads for which nearest neighbor is not meaningful. Often, even the reported experiments, when analyzed carefully, show that linear scan would outperform the techniques being proposed on the workloads studied in high (10-15) dimensionality!.

When is nearest neighbor meaningful

Contrasting treatment‐specific survival using double‐robust estimators

Poor lipid control is a risk factor for cardiovascular diseases and diabetes complications. Frequently, however, patients with these diseases do not achieve blood lipid levels recommended by current standards of care. A retrospective study of 67,244 members eligible for disease management (DM) was initiated to evaluate the ability of interventions to promote improvement in low-density lipoprotein cholesterol (LDL-C) laboratory values for people with cardiovascular diseases or diabetes. The baseline trend in improving LDL-C values in the absence of DM was established. A two-year period prior to the start of the DM intervention was examined to measure the mean percent change in LDL-C values that was occurring in the population. The mean percent change observed for this pre-intervention group was then compared to the change in LDL-C values observed during the DM study period. A significant reduction in elevated LDL-C values (F-test; p < 0.0001) was observed for members who participated in the DM intervention...

Improvement of LDL-C laboratory values achieved by participation in a cardiac or diabetes disease management program.

OBJECTIVE The aim of this study was to examine the direct and mediated effects of a telephonic health coaching program on changes to healthy behaviors, life satisfaction, and optimism. METHODS This longitudinal correlational study of 4881 individuals investigated simple and mediated relationships between participation in a telephonic health risk coaching program and outcomes from three annual Well-being Assessments. RESULTS Program participation was directly related to improvements in healthy behaviors, life satisfaction and optimism, and indirect effects of coaching on these variables concurrently and over a one-year time lag were also supported. CONCLUSIONS Given previous research that improvements to life satisfaction, optimism, and health behaviors are valuable for individuals, employers, and communities, a clearer understanding of intervention approaches that may impact these outcomes simultaneously can drive greater program effectiveness and value on investment.

Direct and Mediated Relationships Between Participation in a Telephonic Health Coaching Program and Health Behavior, Life Satisfaction, and Optimism.

Avoiding Randomization Failure in Program Evaluation

Well-being is linked to important societal factors such as health care costs and productivity and has experienced a surge in development activity of both theories and measurement. This study builds on validation of the Well-Being 5 survey and for the first time applies Item Response Theory, a modern and flexible measurement paradigm, to form the basis of adaptive population well-being measurement. Adaptive testing allows survey questions to be administered selectively, thereby reducing the number of questions required of the participant. After the graded response model was fit to a sample of size N = 12,035, theta scores were estimated based on both the full-item bank and a simulation of Computerized Adaptive Testing (CAT). Comparisons of these 2 sets of score estimates with each other and of their correlations with external outcomes of job performance, absenteeism, and hospital admissions demonstrate that the CAT well-being scores maintain accuracy and validity. The simulation indicates that the...

Carter Coberley

Papers

Improvement of LDL-C laboratory values achieved by participation in a cardiac or diabetes disease management program.

Direct and Mediated Relationships Between Participation in a Telephonic Health Coaching Program and Health Behavior, Life Satisfaction, and Optimism.

Avoiding Randomization Failure in Program Evaluation

Adaptive Measurement of Well-Being: Maximizing Efficiency and Optimizing User Experience during Individual Assessment.

Identification of Gaps in Heart Failure Medication Compliance in an Unintervened Commercial Health Care Population