Mathematics
8 years ago

New method uses generated data to test for normality accurately.

A New method for Testing Normality based upon a Characterization of the Normal Distribution
Davayne A Melbourne

Paper Summary

The article introduces a new method for testing if data follows a normal distribution. Instead of only using observed data, this method combines observed and generated data. The test statistic D* is calculated based on linear combinations of the data. This approach is different from traditional methods and is based on a theorem by Bernstein.

Related papers

Theme background image
Mathematics
8 years ago

No one-size-fits-all: Study reveals best normality test for data shape

Skew Distributions
Model-Based Clustering
Financial Volatility
Theme background image
Computer Science
8 years ago

New Test Accurately Identifies Normality, Revolutionizing Data Analysis

Model-Based Clustering
Multicollinearity
Skew Distributions
Theme background image
Mathematics
8 years ago

New tests for normality could revolutionize data analysis in all fields!

Skew Distributions
Financial Volatility
Model-Based Clustering