Uncertainty Quantification Techniques in Statistics
- MDPI - Multidisciplinary Digital Publishing Institute 2020
- 1 electronic resource (128 p.)
Open Access
Uncertainty quantification (UQ) is a mainstream research topic in applied mathematics and statistics. To identify UQ problems, diverse modern techniques for large and complex data analyses have been developed in applied mathematics, computer science, and statistics. This Special Issue of Mathematics (ISSN 2227-7390) includes diverse modern data analysis methods such as skew-reflected-Gompertz information quantifiers with application to sea surface temperature records, the performance of variable selection and classification via a rank-based classifier, two-stage classification with SIS using a new filter ranking method in high throughput data, an estimation of sensitive attribute applying geometric distribution under probability proportional to size sampling, combination of ensembles of regularized regression models with resampling-based lasso feature selection in high dimensional data, robust linear trend test for low-coverage next-generation sequence data controlling for covariates, and comparing groups of decision-making units in efficiency based on semiparametric regression.
Kullback–Leibler divergence geometric distribution accuracy AUROC allele read counts mixture model low-coverage entropy gene-expression data SCAD data envelopment analysis LASSO high-throughput sandwich variance estimator adaptive lasso semiparametric regression ?1 lasso Laplacian matrix elastic net feature selection sea surface temperature gene expression data Skew-Reflected-Gompertz distribution lasso next-generation sequencing BH-FDR stochastic frontier model ?2 ridge geometric mean resampling Gompertz distribution adapative lasso group efficiency comparison sensitive attribute MCP probability proportional to size (PPS) sampling randomization device SIS Yennum et al.’s model ensembles