Time to Diagnosis: Accounting for Differential Endpoint Follow-up in Multi-Cohort Studies

Cox regression is widely used to analyze discrete survival time data. Differential endpoint follow-up across sub-cohorts where distribution of a covariate varies may cause typical estimators to be biased or inefficient. We demonstrate that with Cardiovascular Health Study data for incident type 2 diabetes. Two cohorts with extremely different race distribution have differential follow-up for fasting glucose levels. We study various scenarios of Cox regression. We suggest an alternative approach, Poisson generalized estimating equations with an offset to accommodate the differential follow-up. We use simulations to contrast the methods.     

Semiparametric log-linear regression for longitudinal measurements subject to outcome-dependent follow-up

A common problem for longitudinal data analyses is that subjects follow-up is irregular, often related to the past outcome or other factors associated with the outcome measure that are not included in the regression model. Analyses unadjusted for outcome-dependent follow-up yield biased estimates. We propose a longitudinal data analysis that can provide consistent estimates in regression models that are subject to outcome-dependent follow-up. We focus on semiparametric marginal log-link regression with arbitrary unspecified baseline function. Based on estimating equations, the proposed class of estimators are root n consistent and asymptotically normal. We present simulation studies that assess the performance of the estimators under finite samples. We illustrate our approach using data from a health services research study.     

Modelling the travel time of transit vehicles in real‐time through a GTFS‐based road network using GPS vehicle locations

Predicting the arrival time of a transit vehicle involves not only knowledge of its current position and schedule adherence, but also traffic conditions along the remainder of the route. Road networks are dynamic and can quickly change from free‐flowing to highly congested, which impacts the arrival time of transit vehicles, particularly buses which often share the road with other vehicles, so reliable predictions need to account for real‐time and future traffic conditions. The first step in this process is to construct a framework with which road state (traffic conditions) can be estimated using real‐time transit vehicle position data. Our proposed framework implements a vehicle model using a particle filter to estimate road travel times, which are used in a second model to estimate real‐time traffic conditions. Although development and testing took place in Auckland, New Zealand, we generalised each component to make the framework compatible with other public transport systems around the world. We demonstrate the real‐time feasibility and performance of our approach in real‐time, where a combination of R and C++ was used to obtain the necessary performance results. Future work will use these estimated traffic conditions in combination with historical data to obtain reliable arrival time predictions of transit vehicles.     

Estimating prediction error for complex samples

With a growing interest in using non-representative samples to train prediction models for numerous outcomes it is necessary to account for the sampling design that gives rise to the data in order to assess the generalized predictive utility of a proposed prediction rule. After learning a prediction rule based on a non-uniform sample, it is of interest to estimate the rule's error rate when applied to unobserved members of the population. Efron (1986) proposed a general class of covariance penalty inflated prediction error estimators that assume the available training data are representative of the target population for which the prediction rule is to be applied. We extend Efron's estimator to the complex sample context by incorporating Horvitz–Thompson sampling weights and show that it is consistent for the true generalization error rate when applied to the underlying superpopulation. The resulting Horvitz–Thompson–Efron estimator is equivalent to dAIC, a recent extension of Akaike's information criteria to survey sampling data, but is more widely applicable. The proposed methodology is assessed with simulations and is applied to models predicting renal function obtained from the large-scale National Health and Nutrition Examination Study survey. The Canadian Journal of Statistics 48: 204–221; 2020 © 2019 Statistical Society of Canada     

Review of the Stata statistical package

Statistics and Computing -     

Weighted empirical adaptive variance estimators for correlated data regression

Estimating equations based on marginal generalized linear models are useful for regression modelling of correlated data, but inference and testing require reliable estimates of standard errors. We introduce a class of variance estimators based on the weighted empirical variance of the estimating functions and show that an adaptive choice of weights allows reliable estimation both asymptotically and by simulation in finite samples. Connections with previous bootstrap and jackknife methods are explored. The effect of reliable variance estimation is illustrated in data on health effects of air pollution in King County, Washington.     

Tests for Regression Models Fitted to Survey Data

Data from complex surveys are being used increasingly to build the same sort of explanatory and predictive models as those used in the rest of statistics. Unfortunately the assumptions underlying standard statistical methods are not even approximately valid for most survey data. The problem of parameter estimation has been largely solved, at least for routine data analysis, through the use of weighted estimating equations, and software for most standard analytical procedures is now available in the major statistical packages. One notable omission from standard software is an analogue of the likelihood ratio test. An exception is the Rao–Scott test for loglinear models in contingency tables. In this paper we show how the Rao–Scott test can be extended to handle arbitrary regression models. We illustrate the process of fitting a model to survey data with an example from NHANES.     

Effects of accuracy feedback versus monetary contingency on arousal in high and low frequency gamblers

This study examined the effects on arousal of feedback regarding prediction accuracy without monetary gain or loss versus accuracy feedback combined with monetary contingency involved in most gambling. Physiological and subjective arousal was assessed in frequent (n = 35; 16 females) and infrequent gamblers (n = 35; 16 females) during participation in a laboratory gambling game. Both samples consisted of undergraduate student volunteers. In one condition, subjects gambled money on their predictions about a sequence of events, hence receiving both feedback and monetary outcome. In the other condition, only feedback about accuracy was provided, with no money wagered. Arousal was greater for both groups during actual gambling than in the feedback only condition. Across both conditions, frequent gamblers exhibited greater arousal than did infrequent gamblers. There were, however, no significant Group X Condition interactions. These results suggest that actual monetary contingency rather than accuracy feedback alone is most motivating in gambling. Methodological limitations possibly related to the lack of Group X Condition interactions were discussed.     

Longitudinal data analysis for generalized linear models with follow‐up dependent on outcome‐related variables

In longitudinal studies, observation times are often irregular and subject‐specific. Frequently they are related to the outcome measure or other variables that are associated with the outcome measure but undesirable to condition upon in the model for outcome. Regression analyses that are unadjusted for outcome‐dependent follow‐up then yield biased estimates. The authors propose a class of inverse‐intensity rate‐ratio weighted estimators in generalized linear models that adjust for outcome‐dependent follow‐up. The estimators, based on estimating equations, are very simple and easily computed; they can be used under mixtures of continuous and discrete observation times. The predictors of observation times can be past observed outcomes, cumulative values of outcome‐model covariates and other factors associated with the outcome. The authors validate their approach through simulations and they illustrate it using data from a supported housing program from the US federal government.