论文信息 - Bias, Variance and Prediction Error for Classification Rules

Bias, Variance and Prediction Error for Classification Rules

We study the notions of bias and variance for classiication rules. Following Efron (1978) we develop a decomposition of prediction error into its natural components. Then we derive bootstrap estimates of these components and illustrate how they can be used to describe the error behaviour of a classiier in practice. In the process we also obtain a bootstrap estimate of the error of a \bagged" classiier.

Robert Tibshirani | R. Tibshirani

[1] Peter E. Hart,et al. Nearest neighbor pattern classification , 1967, IEEE Trans. Inf. Theory.

[2] B. Efron. Regression and ANOVA with Zero-One Data: Measures of Residual Variation , 1978 .

[3] R. Tibshirani,et al. Cross-Validation and the Bootstrap : Estimating the Error Rate ofa Prediction , 1995 .

[4] Leo Breiman,et al. Bias, Variance , And Arcing Classifiers , 1996 .

[5] Ron Kohavi,et al. Bias Plus Variance Decomposition for Zero-One Loss Functions , 1996, ICML.