Mathematics
8 years ago

Leaving-one-out method found best for error rates in analysis

Developments in marketing science: proceedings of the Academy of Marketing Science
A Comparison of the Different Measures of Error Rates in Discriminant Analysis using Small Samples
T. H. Mattheiss, Subhash Sharma

Paper Summary

Related papers

Theme background image
Computer Science
4 years ago

New face recognition technology enhances accuracy and security for all.

Dimensionality Reduction
Hyperspectral Analysis
2 views
Theme background image
Computer Science
5 years ago

New hybrid method revolutionizes machine learning with improved classification accuracy.

Dimensionality Reduction
Control Systems
Neural Networks
Business, Management and Accounting
8 years ago

New method boosts accuracy in credit evaluation of corporations.

Cross-Border E-Commerce
Emergency Management
Ideological Education

The article compares different ways to measure error rates in a type of analysis called discriminant analysis. The researchers used sales data to compare these measures. They found that the leaving-one-out method was the most accurate way to measure error rates.