Weighted Estimating Equations for Additive Hazards Models with Missing Covariates.

This paper presents simple weighted and fully augmented weighted estimators for the additive hazards model with missing covariates when they are missing at random. The additive hazards model estimates the difference in hazards and has an intuitive biological interpretation. The proposed weighted estimators for the additive hazards model use incomplete data nonparametrically and have close-form expressions. We show that they are consistent and asymptotically normal, and are more efficient than the simple weighted estimator which only uses the complete data. We illustrate their finite-sample performance through simulation studies and an application to study the progression from mild cognitive impairment to dementia using data from the Alzheimer's Disease Neuroimaging Initiative as well as an application to the mouse leukemia study.

Mammographic non-dense area and breast cancer risk in postmenopausal women: a causal inference approach in a case-control study.

PURPOSE: The association between high mammographic density (MD) and elevated breast cancer risk is well established. However, the role of absolute non-dense area remains unclear. We estimated the effect of the mammographic non-dense area and other density parameters on the risk of breast cancer. METHODS: This study utilizes data from a population-based case-control study conducted in Greater Vancouver, British Columbia, with 477 female postmenopausal breast cancer cases and 588 female postmenopausal controls. MD measures were determined from digitized screening mammograms using computer-assisted software (Cumulus). Marginal odds ratios were estimated by inverse-probability weighting using a causal diagram for confounder selection. Akaike information criteria and receiver operating characteristic curves were used to assess the goodness of fit and predictive power of unconditional logistic models containing MD parameters. RESULTS: The risk of breast cancer is 60% lower for the highest quartile compared to the lowest quartile of mammographic non-dense area (marginal OR 0.40, 95% CI 0.26-0.61, p-trend < 0.001). The cancer risk almost doubles for the highest quartile compared to the lowest quartile of dense area (marginal OR 1.81, 95% CI 1.19-2.43, p-trend < 0.001). For the highest quartile of percent density, breast cancer risk was more than three times higher than for the lowest quartile (marginal OR 3.15, 95% CI 1.90-4.40, p-trend < 0.001). No difference was seen in predictive accuracy between models using percent density alone, dense area alone, or non-dense area plus dense area. CONCLUSIONS: In this study, non-dense area is an independent risk factor after adjustment for dense area and other covariates, inversely related with the risk of breast cancer. However, non-dense area does not improve prediction over that offered by percent density or dense area alone.

Assessing effects of cholera vaccination in the presence of interference.

Interference occurs when the treatment of one person affects the outcome of another. For example, in infectious diseases, whether one individual is vaccinated may affect whether another individual becomes infected or develops disease. Quantifying such indirect (or spillover) effects of vaccination could have important public health or policy implications. In this article we use recently developed inverse-probability weighted (IPW) estimators of treatment effects in the presence of interference to analyze an individually-randomized, placebo-controlled trial of cholera vaccination that targeted 121,982 individuals in Matlab, Bangladesh. Because these IPW estimators have not been employed previously, a simulation study was also conducted to assess the empirical behavior of the estimators in settings similar to the cholera vaccine trial. Simulation study results demonstrate the IPW estimators can yield unbiased estimates of the direct, indirect, total, and overall effects of vaccination when there is interference provided the untestable no unmeasured confounders assumption holds and the group-level propensity score model is correctly specified. Application of the IPW estimators to the cholera vaccine trial indicates the presence of interference. For example, the IPW estimates suggest on average 5.29 fewer cases of cholera per 1000 person-years (95% confidence interval 2.61, 7.96) will occur among unvaccinated individuals within neighborhoods with 60% vaccine coverage compared to neighborhoods with 32% coverage. Our analysis also demonstrates how not accounting for interference can render misleading conclusions about the public health utility of vaccination.

Unbiased weighted variance and skewness estimators for overlapping returns.

This article develops unbiased weighted variance and skewness estimators for overlapping return distributions. These estimators extend the variance estimation methods constructed in Bod et. al. (Applied Financial Economics 12:155-158, 2002) and Lo and MacKinlay (Review of Financial Studies 1:41-66, 1988). In addition, they may be used in overlapping return variance or skewness ratio tests as in Charles and Darné (Journal of Economic Surveys 3:503-527, 2009) and Wong (Cardiff Economics Working Papers, 2016). An example using synthetic overlapping returns from a model fit to data from the SPY S&P 500 exchange traded fund is given in order to demonstrate under which circumstances the unbiased correction becomes significant in skewness estimation. Finally, we compare the effect of the HAC weighting schemes of Andrews (Econometrica 53:817-858, 1991) as a function of sample size and overlapping return window length.