A synopsis of resampling techniques

作者:Brunelli Alessandro*
来源:Journal of Thoracic Disease, 2014, 6(12): 1882-1885.
DOI:10.3978/j.issn.2072-1439.2014.09.09

摘要

Bootstrap is a computer intensive technique of resampling with replacement, which can be applied in many statistical analytical tests. The article describes the most frequent situations where bootstrap resampling can be applied in thoracic surgical research: variable selection for multivariable regression analysis, internal validation of regression equations, model validation. Practical examples for programming bootstrap in commercially available statistical software are finally reported.

  • 出版日期2014-12