General Pivotal Goodness of Fit Test Based on Kernel Density Estimation
Hani M. Samawi, Robert L. Vogel
Abstract
Hani M. Samawi, Robert L. Vogel
Abstract
In this paper, we introduce a pivotal goodness of fit test based on empirical kernel density estimation. Our investigation reveals that the new test is more powerful than the traditional goodness of tests found in the literature; namely, the Chi-square and the Kolmogorov-Smirnov (KS) goodness of fit tests. Intensive simulation is conducted to examine the power of the proposed test. Data from a level I Trauma center are used to illustrate the procedures developed in this paper.
A significance statement is not available in the OpenAlex record.
A contribution statement is not available in the OpenAlex record.
Method details are not available in the OpenAlex metadata.
Findings are not separately available in the OpenAlex metadata.
Limitations are not available in the OpenAlex metadata.
Application details are not available in the OpenAlex metadata.
In this paper, we introduce a pivotal goodness of fit test based on empirical kernel density estimation. Our investigation reveals that the new test is more powerful than the traditional goodness of tests found in the literature; namely, the Chi-square and the Kolmogorov-Smirnov (KS) goodness of fit tests. Intensive simulation is conducted to examine the power of the proposed test. Data from a level I Trauma center are used to illustrate the procedures developed in this paper.
Key concepts: Goodness of fit, Kernel density estimation, Statistics, Kolmogorov–Smirnov test, Kernel (algebra), Mathematics, Test (biology), Anderson–Darling test