1. Introduction and Overview. 2. Detecting Influential Observations and Outliers. 3. Detecting and Assessing Collinearity. 4. Applications and Remedies. 5. Research Issues and Directions for Extensions. Bibliography. Author Index. Subject Index.

Regression Diagnostics: Identifying Influential Data and Sources of Collinearity

The propensity score is a subject's probability of treatment, conditional on observed baseline covariates. Conditional on the true propensity score, treated and untreated subjects have similar distributions of observed baseline covariates. Propensity-score matching is a popular method of using the propensity score in the medical literature. Using this approach, matched sets of treated and untreated subjects with similar values of the propensity score are formed. Inferences about treatment effect made using propensity-score matching are valid only if, in the matched sample, treated and untreated subjects have similar distributions of measured baseline covariates. In this paper we discuss the following methods for assessing whether the propensity score model has been correctly specified: comparing means and prevalences of baseline characteristics using standardized differences; ratios comparing the variance of continuous covariates between treated and untreated subjects; comparison of higher order moments and interactions; five-number summaries; and graphical methods such as quantile–quantile plots, side-by-side boxplots, and non-parametric density plots for comparing the distribution of baseline covariates between treatment groups. We describe methods to determine the sampling distribution of the standardized difference when the true standardized difference is equal to zero, thereby allowing one to determine the range of standardized differences that are plausible with the propensity score model having been correctly specified. We highlight the limitations of some previously used methods for assessing the adequacy of the specification of the propensity-score model. In particular, methods based on comparing the distribution of the estimated propensity score between treated and untreated subjects are uninformative. Copyright © 2009 John Wiley & Sons, Ltd.

Balance diagnostics for comparing the distribution of baseline covariates between treatment groups in propensity-score matched samples

The 2008 crash has left all the established economic doctrines - equilibrium models, real business cycles, disequilibria models - in disarray. Part of the problem is due to Smith’s "veil of ignorance": individuals unknowingly pursue society’s interest and, as a result, have no clue as to the macroeconomic effects of their actions: witness the Keynes and Leontief multipliers, the concept of value added, fiat money, Engel’s law and technical progress, to name but a few of the macrofoundations of microeconomics. A good viewpoint to take bearings anew lies in comparing the post-Great Depression institutions with those emerging from Thatcher and Reagan’s economic policies: deregulation, exogenous vs. endoge- nous money, shadow banking vs. Volcker’s Rule. Very simply, the banks, whose lending determined deposits after Roosevelt, and were a public service became private enterprises whose deposits determine lending. These underlay the great moderation preceding 2006, and the subsequent crash.

The Economic Institutions of Capitalism

We investigate conditions sufficient for identification of average treatment effects using instrumental variables. First we show that the existence of valid instruments is not sufficient to identify any meaningful average treatment effect. We then establish that the combination of an instrument and a condition on the relation between the instrument and the participation status is sufficient for identification of a local average treatment effect for those who can be induced to change their participation status by changing the value of the instrument. Finally we derive the probability limit of the standard IV estimator under these conditions. It is seen to be a weighted average of local average treatment effects.

Evolution and Rationality Some Recent Game-Theoretic Results. Identification and Estimation of Local Average Treatment Effects

Statistical Analysis with Missing Data

https://www.fertstert.org/article/S0015-0282(03)00345-5/pdf

Local injury to the endometrium doubles the incidence of successful pregnancies in patients undergoing in vitro fertilization.

https://www.fertstert.org/article/S0015-0282(10)00272-4/pdf

Local injury of the endometrium induces an inflammatory response that promotes successful implantation

Background: Although smoking is a major cause of lung cancer, only a proportion of smokers develop lung cancer, suggesting a genetic predisposition in some individuals. Because tobacco smoking is associated with the increased formation of DNA lesions, including those induced from oxidative damage, we investigated whether the activity of the DNA repair enzyme 8-oxoguanine DNA N-glycosylase (OGG), which repairs the oxidative DNA lesion 8-oxoguanine, is associated with lung cancer. Methods: We conducted a molecular epidemiologic case–control study that included 68 case patients with non–small-cell lung cancer and 68 healthy control subjects, frequency matched for age and sex. Enzymatic OGG activity was determined in protein extracts prepared from peripheral blood mononuclear cells or lung tissue by assaying the cleavage product of a radiolabeled synthetic DNA oligonucleotide containing an 8-oxoguanine residue. Odds ratios (ORs) and 95% confidence intervals (CIs) were determined by conditional logistic regression. All statistical tests were two-sided. Results: OGG activity was lower in peripheral blood mononuclear cells from case patients than in those from control subjects. After adjustment for age and smoking status, individuals in the lowest tertile of OGG activity had an increased risk of non–small-cell lung cancer compared with individuals in the highest tertile (OR = 4.8, 95% CI = 1.5 to 15.9). The adjusted OR associated with a unit decrease in OGG activity was statistically significantly increased (OR = 1.9, 95% CI = 1.3 to 2.8). There was no interaction between OGG activity and smoking status. The estimated relative risk of lung cancer for smokers with low OGG activity was 34- or 124-fold higher for smokers with a low OGG activity of 6.0 or 4.0 U/µg protein, respectively, than for nonsmokers with a normal OGG activity of 7.0 U/µg protein, illustrating the cumulative effect of low OGG activity and smoking. Conclusions: Low OGG activity is associated with an increased risk of lung cancer. Although prospective studies are needed to validate the results, they suggest that smoking cessation in individuals with reduced OGG activity might be an effective strategy in lung cancer prevention. [J Natl Cancer Inst 2003;95:1312–19]

/pdf/dna-repair-activity-for-oxidative-damage-and-risk-of-lung-etrr7k8slh.pdf

DNA Repair Activity for Oxidative Damage and Risk of Lung Cancer

Self-reports of safe driving behaviors in relationship to sex, age, education and income in the US adult driving population.

On donne deux idees pour renforcer une simulation: la premiere est d'equilibrer les echantillons simules, la seconde est de faire un usage explicite des approximations qui ne demandent pas de simulation

Edna Schechtman

Papers

Local injury to the endometrium doubles the incidence of successful pregnancies in patients undergoing in vitro fertilization.

Local injury of the endometrium induces an inflammatory response that promotes successful implantation

DNA Repair Activity for Oxidative Damage and Risk of Lung Cancer

Self-reports of safe driving behaviors in relationship to sex, age, education and income in the US adult driving population.

Efficient bootstrap simulation