Nonparametric Statistical Inference

Gibbons, Jean Dickinson; Chakraborti, S.

doi:10.1201/9781439896129

Cited by 433 publications

(395 citation statements)

References 0 publications

Supporting

Mentioning

376

Contrasting

Unclassified

Order By: Relevance

“…This test has been used to analyze supply chain performance in Donohue (2003, 2006), Wu and Katok (2006), Macdonald (2009), andCroson et al (2014). See Mann and Whitney (1947) and Gibbons (1971) for details of this test. 4 We thank an anonymous referee for this suggestion.…”

Section: Resultsmentioning

confidence: 99%

Decision Making and Cognition in Multi‐Echelon Supply Chains: An Experimental Study

Narayanan

Moritz

2015

Production and Operations Management

View full text Add to dashboard Cite

Supply chain performance often depends on the individual decisions of channel members. Even when individuals have access to relevant information, order variation tends to increase when moving up the supply chain, a phenomenon known as the bullwhip effect. While prior research has investigated several structural/environmental factors which can mitigate the bullwhip effect, the underlying behavioral factors contributing to it are an open question. Using a production and distribution decision‐making simulation representing a four‐stage serial supply chain, we find that the cognitive profile of decision makers contributes to the bullwhip effect. We found that the specific decision tendency to underweight the supply line is linked to an individual's level of cognitive reflection. Furthermore, performance differs for entire supply chains and for specific echelons, and holds under standard mitigation efforts. The findings have implications for supply chain design, education, and industry.

show abstract

Section: Resultsmentioning

confidence: 99%

Decision Making and Cognition in Multi‐Echelon Supply Chains: An Experimental Study

Narayanan

Moritz

2015

Production and Operations Management

View full text Add to dashboard Cite

show abstract

“…Adequate power considerations were based on a table of critical values of Spearman's r s for small sample size. (17) Table values indicate that at the 10% significance level for a sample size of 5, the critical value for r s is 0.900. The small sample size for this study places it in the category of being a pilot effort.…”

Section: Methodsmentioning

confidence: 99%

Reducing Risk on Machinery: A Field Evaluation Pilot Study of Risk Assessment

Etherton¹,

Main²,

Cloutier³

et al. 2008

Risk Analysis

View full text Add to dashboard Cite

A pilot evaluation of the ANSI B11-TR3 Machinery Risk Assessment/Risk Reduction (RA/RR) Guideline was conducted. The TR3 guideline was introduced into five companies on one machinery system in each company with a second machine system serving as a control. A pre-post investigation was performed with safety conditions measured pre and post in both treatment and control and with risk reduction score measured only in the treatment machine system. NIOSH provided a commercially available risk assessment software to facilitate the process. Evaluation measures included avoided injuries, reduced exposure to machinery hazards, pretest and posttest knowledge demonstration, assessment of group processes following training, correct implementation of the guidelines, and degree to which risk reduction recommendations were implemented. The qualitative results of this pilot effort appear to be the best indicators for the way ahead in industrial machine risk assessment. All companies indicated that they derived value in participating in this study and in conducting risk assessments. Quantitative study results suggest that: (1) as measured by the knowledge of the participants before and after the TR3 training, the guidelines can be effective at enhancing employee knowledge of safe machine operations and (2) although the injury reduction trends appear successful, the small sample size in the study size should be considered in interpreting these early results.

show abstract

“…Here MST(S) denotes the MST graph over nice and finite set S ⊂ R d and η is the smoothness Hölder parameter. Note that ς η (l, k) is given as before in Lemma 4 (27). Theorem 8: Assume R m,n := R(X m , Y n ) denotes the FR test statistic and densities f 0 and f 1 belong to the strong Hölder class Σ S d (η, L), 0 < η ≤ 1.…”

Section: Appendix B Proof Of Theoremmentioning

confidence: 99%

Convergence Rates for Empirical Estimation of Binary Classification Bounds

Sekeh

Noshad

Moon

et al. 2019

Entropy

View full text Add to dashboard Cite

Bounding the best achievable error probability for binary classification problems is relevant to many applications including machine learning, signal processing, and information theory. Many bounds on the Bayes binary classification error rate depend on information divergences between the pair of class distributions. Recently, the Henze-Penrose (HP) divergence has been proposed for bounding classification error probability. We consider the problem of empirically estimating the HP-divergence from random samples. We derive a bound on the convergence rate for the Friedman-Rafsky (FR) estimator of the HP-divergence, which is related to a multivariate runs statistic for testing between two distributions. The FR estimator is derived from a multicolored Euclidean minimal spanning tree (MST) that spans the merged samples. We obtain a concentration inequality for the Friedman-Rafsky estimator of the Henze-Penrose divergence. We validate our results experimentally and illustrate their application to real datasets.

show abstract

Nonparametric Statistical Inference

Cited by 433 publications

References 0 publications

Decision Making and Cognition in Multi‐Echelon Supply Chains: An Experimental Study

Decision Making and Cognition in Multi‐Echelon Supply Chains: An Experimental Study

Reducing Risk on Machinery: A Field Evaluation Pilot Study of Risk Assessment

Convergence Rates for Empirical Estimation of Binary Classification Bounds

Contact Info

Product

Resources

About