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.     

Three's a crowd? The Nader vote in the 2000 presidential, election in U.S. metropolitan areas

The 2000 U.S. presidential election resulted in one of the closest and most controversial outcomes in U.S. history. Green Party candidate Ralph Nader had little chance of winning, but nevertheless impacted the race in several close states and arguably swung the race from Al Gore to George W. Bush. This research examines Ralph Nader's “urban strategy” to win 5% of the vote for president and the bases of his electoral support. This study uses the METRO_2000 data set which contains a variety of publicly available variables for 276 U.S. metropolitan statistical areas (MSAs) in the year 2000. The analysis uses OLS regression to examine the determinants of the percentage of the vote for Nader in each MSA. The results indicate that the Nader vote was positively influenced by key electoral variables such as the level of electoral participation, whether Nader was on the ballot or could be written in, and the closeness of the race in state polls leading up to the election. The Nader vote was also higher in MSAs with high percentages of voters who supported Nader programs including environmentalists, those favoring universal health care and gay rights, union members, and MSAs that were college towns or with high percentages of college-educated voters. Ralph Nader's urban strategy effectively mobilized enough of his electoral base to impact the 2000 election, but electoral constraints and the closeness of the race prevented him from achieving his goal of attaining 5% of the vote. This case holds lessons about the limits and possibilities of third party campaigns in U.S. presidential elections.     

CREDIT,WAGES, AND BANKRUPTCY LAWS

We analyze how bankruptcy laws affect the general equilibrium interactions between credit and wages. Soft laws reduce the frequency of liquidations and thus ex post inefficiencies, but they worsen credit rationing ex ante. This hinders firm creation and thus depresses labor demand. Rich agents who need few outside funds can invest even if creditor rights are weak. Hence, they favor soft laws that exclude poorer agents from the credit market and reduce the competition for labor. Such laws can generate greater utilitarian welfare than under perfect contract enforcement: By barring access to credit to some agents, soft laws lower wages, which increases the pledgeable income of richer agents and decreases the liquidation rates they must commit to. When they induce strong credit rationing, however, soft laws are Pareto‐dominated by tougher laws combined with subsidies to entrepreneurs. (JEL: D82, G33, K22)     

Toward a General Theory of Alienation*

In this paper, the author arranges some of the major treatments of alienation within a coherent framework. He introduces a distinction between disassociation (certain specified objective conditions) and alienation (a specified experience) and articulates the relationship between these two by presenting a series of intervening psychological issues. A typology of human association is presented; and it is argued that especially two types of disassociation (i.e., marginality and subordination) and, to some extent, another (i.e., privilege) may be productive of alienation. These conditions are considered at the cultural, social, and psychological levels of analysis. Variables connecting these conditions and alienation include the perception of the condition, evaluation, integration, blame, and response.     

Unconditional Quantile Regressions

We propose a new regression method to evaluate the impact of changes in the distribution of the explanatory variables on quantiles of the unconditional (marginal) distribution of an outcome variable. The proposed method consists of running a regression of the (recentered) influence function (RIF) of the unconditional quantile on the explanatory variables. The influence function, a widely used tool in robust estimation, is easily computed for quantiles, as well as for other distributional statistics. Our approach, thus, can be readily generalized to other distributional statistics.     

An evaluation of the bootstrap for model validation in mixture models

Bootstrapping has been used as a diagnostic tool for validating model results for a wide array of statistical models. Here we evaluate the use of the non-parametric bootstrap for model validation in mixture models. We show that the bootstrap is problematic for validating the results of class enumeration and demonstrating the stability of parameter estimates in both finite mixture and regression mixture models. In only 44% of simulations did bootstrapping detect the correct number of classes in at least 90% of the bootstrap samples for a finite mixture model without any model violations. For regression mixture models and cases with violated model assumptions, the performance was even worse. Consequently, we cannot recommend the non-parametric bootstrap for validating mixture models.

The cause of the problem is that when resampling is used influential individual observations have a high likelihood of being sampled many times. The presence of multiple replications of even moderately extreme observations is shown to lead to additional latent classes being extracted. To verify that these replications cause the problems we show that leave-k-out cross-validation where sub-samples taken without replacement does not suffer from the same problem.     

Computer experiment designs for accurate prediction

Computer experiments using deterministic simulators are sometimes used to replace or supplement physical system experiments. This paper compares designs for an initial computer simulator experiment based on empirical prediction accuracy; it recommends designs for producing accurate predictions. The basis for the majority of the designs compared is the integrated mean squared prediction error (IMSPE) that is computed assuming a Gaussian process model with a Gaussian correlation function. Designs that minimize the IMSPE with respect to a fixed set of correlation parameters as well as designs that minimize a weighted IMSPE over the correlation parameters are studied. These IMSPE-based designs are compared with three widely-used space-filling designs. The designs are used to predict test surfaces representing a range of stationary and non-stationary functions. For the test conditions examined in this paper, the designs constructed under IMSPE-based criteria are shown to outperform space-filling Latin hypercube designs and maximum projection designs when predicting smooth functions of stationary appearance, while space-filling and maximum projection designs are superior for test functions that exhibit strong non-stationarity.