On approximate robust confidence distributions

Approximate confidence interval for standard deviation of nonnormal distributions

The exact confidence interval for is hypersensitive tominor violations of the normality assumption and its performance does not improvewith increasing sample size.An approximate confidence interval for is proposed and is shown to be nearly exact under normality with excellent small-sample properties under moderate nonnormality. The small-sample performance of the proposed interval may be furthe...

Confidence Interval Distributions 1

Methodological recommendations strongly emphasize the routine reporting of effect sizes and associated confidence intervals to express the uncertainty around the primary outcomes. Confidence intervals (CI) for unstandardized effects are easy to construct; However, CI’s for standardized measures such as the standardized mean difference (i.e., Cohen’s d), the Pearson product-moment correlation, p...

Confidence Intervals Based on Robust Estimators

Robust semantic confidence scoring

This paper describes an approach for defining robust, application-independent confidence measures for dialogue systems. A concept-level confidence score is computed using a Multi-Layer Perceptron (MLP) classifier trained to discriminate between correct and incorrect concepts. Three types of concept-level confidence features are considered: features based on the confidence score of the underlyin...

Robust approximate assembly partitioning

We present a robust approximate assembly partitioning algorithm for polyhedral parts. We achieve robustness by applying our controlled linear perturbation strategy to Minkowski sums of polyhedra and to arrangements of great circle arcs. Our algorithm is far faster than a prior robust algorithm based on exact computational geometry. Its error is small even on degenerate input.