How Biased is the Apparent Error Rate of a Prediction Rule? (Q3757198)

From MaRDI portal

!

This is the item page for this Wikibase entity, intended for internal use and editing purposes. Please use the normal view instead:

scientific article; zbMATH DE number 4007505
Language Label Description Also known as
default for all languages
No label defined
    English
    How Biased is the Apparent Error Rate of a Prediction Rule?
    scientific article; zbMATH DE number 4007505

      Statements

      0 references
      1986
      0 references
      downward bias
      0 references
      general exponential family linear models
      0 references
      AIC
      0 references
      underestimation
      0 references
      optimism
      0 references
      error rate of prediction rule
      0 references
      prediction errors
      0 references
      prediction model
      0 references
      generalized linear model
      0 references
      logistic model
      0 references
      Numerical results
      0 references
      example
      0 references
      measures of prediction errors
      0 references
      Mallow's Cp
      0 references
      cross validation
      0 references
      bootstrap
      0 references
      Akaike criterium
      0 references
      How Biased is the Apparent Error Rate of a Prediction Rule? (English)
      0 references
      The considered problem is to correct the underestimation, so called optimism, of the error rate of prediction rule. This optimism is a consequence of the fact that the error rate is calculated from the prediction errors of the observations which are used to select the prediction model. The general expression of the optimism is related to the prediction rule. The case of the generalized linear model is considered. Special attention is given to the logistic model. Numerical results concerning an example, using several measures of prediction errors (Mallow's Cp, cross validation, bootstrap and Akaike criterium) are given.
      0 references

      Identifiers

      0 references
      0 references
      0 references
      0 references
      0 references
      0 references
      0 references