Biometrika Centenary: Nonparametrics

• 成果类型: Review
• 署名作者: Hall, P
• 署名单位: Australian National University
• 刊物名称: BIOMETRIKA
• ISSN/ISSBN: 0006-3444
• DOI: 10.1093/biomet/88.1.143
• 发表日期: 2001
• 页码: 143165
• 关键词: bivariate survival function; kernel density-estimation; maximum-likelihood estimator; interval censored-data; Empirical Likelihood; confidence-intervals; Cross-validation; variance-estimation; monte-carlo; bandwidth selection

摘要

Contributions in Biometrika to nonparametric statistics are surveyed, with particular emphasis on more recent work. It is argued that even in the journal's very early years it had interests in smoothing methods for density estimation, and in techniques for nonparametric regression, in addition to its more obvious focus on 'classical' nonparametric statistics such as ranking methods. Biometrika's early involvement with permutation testing has links to its contemporary work on bootstrap methods. Emphases on different types of nonparametric statistics have evolved over time, however. A rapidly increasing proportion of the journal's pages is devoted to papers that include some use of nonparametric ideas or methods.