Conditional prediction intervals for linear regression

被引:5
|
作者
McCullagh, Peter [1 ]
Vovk, Vladimir [2 ]
Nouretdinov, Ilia [2 ]
Devetyarov, Dmitry [2 ]
Gammerman, Alex [2 ]
机构
[1] Univ Chicago, Dept Stat, 5734 Univ Ave, Chicago, IL 60637 USA
[2] Royal Holloway Univ London, Dept Comp Sci, Comp Learning Res Ctr, Egham TW20 0EX, Surrey, England
基金
英国工程与自然科学研究理事会; 美国国家科学基金会;
关键词
D O I
10.1109/ICMLA.2009.115
中图分类号
TP18 [人工智能理论];
学科分类号
081104 ; 0812 ; 0835 ; 1405 ;
摘要
We construct prediction intervals for the linear regression model with IID errors with a known distribution, not necessarily Gaussian. The coverage probability of our prediction intervals is equal to the nominal confidence level not only unconditionally but also conditionally given a natural a-algebra of invariant events. This implies, in particular, the perfect calibration of our prediction intervals in the online mode of prediction.
引用
收藏
页码:131 / +
页数:2
相关论文
共 50 条