Coverage probability

About: Coverage probability is a research topic. Over the lifetime, 2479 publications have been published within this topic receiving 53259 citations.

Confidence sets based on penalized maximum likelihood estimators in Gaussian regression

Abstract: Confidence intervals based on penalized maximum likelihood estimators such as the LASSO, adaptive LASSO, and hard-thresholding are analyzed. In the known-variance case, the finite-sample coverage properties of such intervals are determined and it is shown that symmetric intervals are the shortest. The length of the shortest intervals based on the hard-thresholding estimator is larger than the length of the shortest interval based on the adaptive LASSO, which is larger than the length of the shortest interval based on the LASSO, which in turn is larger than the standard interval based on the maximum likelihood estimator. In the case where the penalized estimators are tuned to possess the ‘sparsity property’, the intervals based on these estimators are larger than the standard interval by an order of magnitude. Furthermore, a simple asymptotic confidence interval construction in the ‘sparse’ case, that also applies to the smoothly clipped absolute deviation estimator, is discussed. The results for the known-variance case are shown to carry over to the unknown-variance case in an appropriate asymptotic sense.

Comments on construction of confidence intervals in connection with partial least squares

Abstract: Faber and Kowalski recently proposed a method to calculate confidence regions for the regression coefficients in a linear model when partial least squares (PLS) has been used as an estimation method (J. Chemometrics, 11, 181 (1997)). In this short communication we show that the proposed confidence regions fail to have the correct coverage probability. The reason for this is that the PLS estimator is generally biased, whereas the method proposed by Faber and Kowalski requires an unbiased estimator. Similar objections made against PLS also apply to the second regression method mentioned by Faber and Kowalski, principal component regression (PCR). © 1998 John Wiley & Sons, Ltd.

HDRE: Coverage hole detection with residual energy in wireless sensor networks

Abstract: Coverage completeness is an important indicator for quality of service in wireless sensor networks (WSN). Due to limited energy and diverse working conditions, the sensor nodes have different lifetimes which often cause network holes. Most of the existing methods expose large limitation and one-sidedness because they generally consider only one aspect, either coverage rate or energy issue. This paper presents a novel method for coverage hole detection with residual energy in randomly deployed wireless sensor networks. By calculating the life expectancy of working nodes through residual energy, we make a trade-off between network repair cost and energy waste. The working nodes with short lifetime are screened out according to a proper ratio. After that, the locations of coverage holes can be determined by calculating the joint coverage probability and the evaluation criteria. Simulation result shows that compared to those traditional algorithms without consideration of energy problem, our method can effectively maintain the coverage quality of repaired WSN while enhancing the life span of WSN at the same time.

Asymptotic simultaneous confidence intervals for many-to-one comparisons of binary proportions in randomized clinical trials.

Abstract: The simultaneous comparison of proportions of success between many treatments and one control group is a common problem in randomized clinical trials or toxicity studies. In this article, three recently recommended asymptotic confidence interval approaches for the difference of proportions are adjusted for multiplicity, taking the correlation into account. The coverage probability of the resulting interval methods is compared in a simulation study using parameter settings relevant for clinical trials. For moderate to small sample sizes, a method adding two successes and two failures can be recommended. The usage of the proposed methods is illustrated by two examples; an R package is available.