Design and analysis of computer experiments when the output is highly correlated over the input space

To build a predictor, the output of a deterministic computer model or “code” is often treated as a realization of a stochastic process indexed by the code's input variables. The authors consider an asymptotic form of the Gaussian correlation function for the stochastic process where the correlation tends to unity. They show that the limiting best linear unbiased predictor involves Lagrange interpolating polynomials; linear model terms are implicitly included. The authors then develop optimal designs based on minimizing the limiting integrated mean squared error of prediction. They show through several examples that these designs lead to good prediction accuracy.     

Amateur craft‐artists: Marginal leisure roles in a marginal art world (1)

Amateurism is an important but marginal leisure role, in which the boundaries between work and leisure are blurred. This phenomenon has been studied by Stebbins as a subjective social reality. His conceptualization of amateurism was used as a model in the examination of survey data on craft‐artists and it was found to apply rather well to the analysis of quantitative data. As a self‐identified category of craft‐artists, amateurs were statistically marginal between professionals and dabblers.     

On Semiparametric Mode Regression Estimation

It has been found that, for a variety of probability distributions, there is a surprising linear relation between mode, mean, and median. In this article, the relation between mode, mean, and median regression functions is assumed to follow a simple parametric model. We propose a semiparametric conditional mode (mode regression) estimation for an unknown (unimodal) conditional distribution function in the context of regression model, so that any m-step-ahead mean and median forecasts can then be substituted into the resultant model to deliver m-step-ahead mode prediction. In the semiparametric model, Least Squared Estimator (LSEs) for the model parameters and the simultaneous estimation of the unknown mean and median regression functions by the local linear kernel method are combined to infer about the parametric and nonparametric components of the proposed model. The asymptotic normality of these estimators is derived, and the asymptotic distribution of the parameter estimates is also given and is shown to follow usual parametric rates in spite of the presence of the nonparametric component in the model. These results are applied to obtain a data-based test for the dependence of mode regression over mean and median regression under a regression model.     

Clustering objects on subsets of attributes (with discussion)

Summary.  A new procedure is proposed for clustering attribute value data. When used in conjunction with conventional distance-based clustering algorithms this procedure encourages those algorithms to detect automatically subgroups of objects that preferentially cluster on subsets of the attribute variables rather than on all of them simultaneously. The relevant attribute subsets for each individual cluster can be different and partially (or completely) overlap with those of other clusters. Enhancements for increasing sensitivity for detecting especially low cardinality groups clustering on a small subset of variables are discussed. Applications in different domains, including gene expression arrays, are presented.     

On the relative behavior of estimators of reliability/survivability

Based on a decomposition of mean absolute error, a twofold technique is introduced whereby a pairwise comparison of point estimators of reliability/survivability can be made. Given two such estimators, the method examines (a) the “odds” in favor of one of the estimators being closer to the true value than is the other and (b) each estimator’s average closeness to the true value not only when it is closer than is the other but also when it is not. Joint consideration of these concepts is shown to form a basis for determining which of the two estimators is preferred in a given situation. An application of the theory is made by comparing the maximum likelihood and minimum variance unbiased estimators of reliability/survivability in the exponential failure model.     

An adaptive estimatcr frequently used in material fatigue testing

Small sample sizes in material fatigue tests give rise to an adaptive estimator of the 100(1-P)% normal percentile, which is the 100P% survival load. The mean squared relative efficiency of the best invariant estimator of normal percentiles to the adaptive estimator is dependent upon the unknown parameters only through the coefficient of variation. The adaptive estimator is shown to be more locally efficient than tl-:ebest invariant estimator over a subset of the parameter space. However, in the extreme values of P the coverage probabilities of the adaptive estimator provide little more protection than a traditional point estimator over the range of preference based on mean squared relative efficiency.     

On bagging and nonlinear estimation

We propose an elementary model for the way in which stochastic perturbations of a statistical objective function, such as a negative log-likelihood, produce excessive nonlinear variation of the resulting estimator. Theory for the model is transparently simple, and is used to provide new insight into the main factors that affect performance of bagging. In particular, it is shown that if the perturbations are sufficiently symmetric then bagging will not significantly increase bias; and if the perturbations also offer opportunities for cancellation then bagging will reduce variance. For the first property it is sufficient that the third derivative of a perturbation vanish locally, and for the second, that second and fourth derivatives have opposite signs. Functions that satisfy these conditions resemble sinusoids. Therefore, our results imply that bagging will reduce the nonlinear variation, as measured by either variance or mean-squared error, produced in an estimator by sinusoid-like, stochastic perturbations of the objective function. Analysis of our simple model also suggests relationships between the results obtained using different with-replacement and without-replacement bagging schemes. We simulate regression trees in settings that are far more complex than those explicitly addressed by the model, and find that these relationships are generally borne out.     

Occupational Advantage and the Impact of College Rank on Labor Market Outcomes*

Most analyses of the effects of college rank on labor market outcomes focus on its average impact across sub-populations and employment situations. The framework adopted in this paper, however, suggests that the effects of college rank may vary by individual characteristics and type of job. Using data from the 1973 Occupational Changes in a Generation Survey, we test three hypotheses suggesting that occupational advantage, as measured either by family of origin or current position, is a significant determinant of capacity to convert increments in college prestige into labor market success. The findings suggest that where one went to college is especially consequential for (1) the occupational status of individuals from professional families; (2) the earnings of individuals from both professional and managerial families; and (3) the earnings of individuals in both professional and managerial positions. Overall, these findings tend to confirm the study's underlying hypothesis that aggregate analyses of college rank may be misleading when applied across sub-populations and employment situations.     

An Analysis of the Uncertainties in Estimates of Radon-Induced Lung Cancer

A recent report by the National Academy of Sciences estimates that the radiation dose to the bronchial epithelium, per working level month (WLM) of radon daughter exposure, is about 30% lower for residential exposures than for exposures received in underground mines. Adjusting the previously published BEIR IV radon risk model accordingly, the unit risk for indoor exposures of the general population is about 2.2 x 10(-4) lung cancer deaths (lcd)/WLM. Using results from EPA's National Residential Radon Survey, the average radon level is estimated to be about 1.25 pCi/L, and the annual average exposure about 0.242 WLM. Based on these estimates, 13,600 radon-induced lcd/yr are projected for the United States. A quantitative uncertainty analysis was performed, which considers: statistical uncertainties in the epidemiological studies of radon-exposed miners; the dependence of risk on age at, and time since, exposure; the extrapolation of risk estimates from mines to homes based on comparative dosimetry; and uncertainties in the radon daughter levels in homes and in the average residential occupancy. Based on this assessment of the uncertainties in the unit risk and exposure estimates, an uncertainty range of 7000-30000 lcd/yr is derived.     

Small Sample Tests for Shape Parameters of Gamma Distributions

The introduction of shape parameters into statistical distributions provided flexible models that produced better fit to experimental data. The Weibull and gamma families are prime examples wherein shape parameters produce more reliable statistical models than standard exponential models in lifetime studies. In the presence of many independent gamma populations, one may test equality (or homogeneity) of shape parameters. In this article, we develop two tests for testing shape parameters of gamma distributions using chi-square distributions, stochastic majorization, and Schur convexity. The first one tests hypotheses on the shape parameter of a single gamma distribution. We numerically examine the performance of this test and find that it controls Type I error rate for small samples. To compare shape parameters of a set of independent gamma populations, we develop a test that is unbiased in the sense of Schur convexity. These tests are motivated by the need to have simple, easy to use tests and accurate procedures in case of small samples. We illustrate the new tests using three real datasets taken from engineering and environmental science. In addition, we investigate the Bayes’ factor in this context and conclude that for small samples, the frequentist approach performs better than the Bayesian approach.