Home About us Contact
|
Home About us Contact | ||
|
|
||
Practical Setting (practical + setting)
Selected AbstractsTHE SIGNAL DETECTION THEORY ROC CURVE: SOME APPLICATIONS IN FOOD SENSORY SCIENCEJOURNAL OF SENSORY STUDIES, Issue 2 2008M. O'MAHONY ABSTRACT In psychology, the receiver operating characteristic (ROC) curve is a key part of Signal Detection Theory, which is used for calculating d, values in discrimination tests. In food sensory science, the ROC curve can also be a useful tool. To give a specific example, it is not always convenient to use forced-choice protocols for difference tests; foods may be fatiguing, and assessments with single presentations, like the Yes,No procedure, might be more appropriate. In this case, ROC curves provide a useful method for computing d, values. More generally, ROC curves give information about cognitive strategies. Cognitive strategies are important for difference tests. Values of d, can only be computed if the cognitive strategy used in the test is known. PRACTICAL APPLICATIONS When using methods other than two-alternative forced-choice in difference testing, a receiver operating characteristic (ROC) curve would be required to compute d,. This is because when assessing discrimination ability, the cognitive strategy of the subject must be taken into account, and ROC curves can sometimes reveal the cognitive strategy used by the subject. This article describes how the cognitive strategy can be determined from the subject's ROC curve. The hidden assumptions made when using ROC curves and how these assumptions can be tested are also given. This information is essential to researchers in sensory evaluation as well as those using these methods in a practical setting. [source] Recursive algorithms for unbalanced banded Toeplitz systemsNUMERICAL LINEAR ALGEBRA WITH APPLICATIONS, Issue 7 2009P. Favati Abstract Direct recursive algorithms for the solution of band Toeplitz systems are considered here. They exploit the displacement rank properties, which allow a large reduction of computational efforts and storage requirements. Their use of the Sherman,Morrison,Woodbury formula turns out to be particularly suitable for the case of unbalanced bandwidths. The computational costs of the algorithms under consideration are compared both in a theoretical and practical setting. Some stability issues are discussed as well. Copyright © 2009 John Wiley & Sons, Ltd. [source] Maximum likelihood estimation in semiparametric regression models with censored dataJOURNAL OF THE ROYAL STATISTICAL SOCIETY: SERIES B (STATISTICAL METHODOLOGY), Issue 4 2007D. Zeng Summary., Semiparametric regression models play a central role in formulating the effects of covariates on potentially censored failure times and in the joint modelling of incomplete repeated measures and failure times in longitudinal studies. The presence of infinite dimensional parameters poses considerable theoretical and computational challenges in the statistical analysis of such models. We present several classes of semiparametric regression models, which extend the existing models in important directions. We construct appropriate likelihood functions involving both finite dimensional and infinite dimensional parameters. The maximum likelihood estimators are consistent and asymptotically normal with efficient variances. We develop simple and stable numerical techniques to implement the corresponding inference procedures. Extensive simulation experiments demonstrate that the inferential and computational methods proposed perform well in practical settings. Applications to three medical studies yield important new insights. We conclude that there is no reason, theoretical or numerical, not to use maximum likelihood estimation for semiparametric regression models. We discuss several areas that need further research. [source] Conditional Generalized Estimating Equations for the Analysis of Clustered and Longitudinal DataBIOMETRICS, Issue 3 2008Sylvie Goetgeluk Summary A common and important problem in clustered sampling designs is that the effect of within-cluster exposures (i.e., exposures that vary within clusters) on outcome may be confounded by both measured and unmeasured cluster-level factors (i.e., measurements that do not vary within clusters). When some of these are ill/not accounted for, estimation of this effect through population-averaged models or random-effects models may introduce bias. We accommodate this by developing a general theory for the analysis of clustered data, which enables consistent and asymptotically normal estimation of the effects of within-cluster exposures in the presence of cluster-level confounders. Semiparametric efficient estimators are obtained by solving so-called conditional generalized estimating equations. We compare this approach with a popular proposal by Neuhaus and Kalbfleisch (1998, Biometrics54, 638,645) who separate the exposure effect into a within- and a between-cluster component within a random intercept model. We find that the latter approach yields consistent and efficient estimators when the model is linear, but is less flexible in terms of model specification. Under nonlinear models, this approach may yield inconsistent and inefficient estimators, though with little bias in most practical settings. [source] | ||||||||||