Estimationg error rates in discriminant analysis with correlated training observations: a simulation study

This article reports results of an extensive simulation study which investigated the performances of some commonly used methods of estimating error rates in discriminant analysis. Earlier research papers limited their comparisons of these methods to independent training data. This study allows for a simple auto-regressive dependence among the training data. The results suggest that the estimation methods based on the normal distribution perform adequately well under conditions of negative or mild positive correlation in the data, and small dimensions (p) of the observation vectors. For large p or strong positive correlation structures the conclusion is that one of the better non-parametric methods should be used. Special circumstances and conditions which notably affect the relative performances of the methods are identified.

[1]  Patrick L. Odell,et al.  Effect of intraclass correlation among training samples on the misclassification probabilities of bayes procedure , 1974, Pattern Recognit..

[2]  R. G. Craig,et al.  Autocorrelation in Landsat data , 1979 .

[3]  Jake D. Tubbs,et al.  Effect of autocorrelated training samples on Bayes' probabilities of misclassification , 1980, Pattern Recognit..

[4]  B. Efron.  Bootstrap Methods: Another Look at the Jackknife , 1979 .

[5]  Geoffrey J. McLachlan,et al.  Some asymptotic results on the effect of autocorrelation on the error rates of the sample linear discriminant function , 1983, Pattern Recognit..

[6]  G. McLachlan.  Estimation of the Errors of Misclassification on the Criterion of Asymptotic Mean Square Error , 1974 .

[7]  S. Snapinn,et al.  An Evaluation of Smoothed Classification Error- Rate Estimators , 1985 .

[8]  James D. Knoke,et al.  Bootstrapped and smoothed classification error rate estimators , 1988 .

[9]  J. Page.  Error-Rate Estimation in Discriminant Analysis , 1985 .

[10]  D. J. Hand,et al.  Recent advances in error rate estimation , 1986, Pattern Recognit. Lett..

[11]  M. Okamoto.  An Asymptotic Expansion for the Distribution of the Linear Discriminant Function , 1963 .

[12]  G. McLachlan.  Error Rate Estimation in Discriminant Analysis: Recent Advances , 1987 .

[13]  Wojtek J. Krzanowski,et al.  Error-rate estimation in two-group discriminant analysis using the linear discriminant function , 1990 .

[14]  G. McLachlan.  An Asymptotic Unbiased Technique for Estimating the Error Rates in Discriminant Analysis , 1974 .

[15]  G. McLachlan.  Discriminant Analysis and Statistical Pattern Recognition , 1992 .

[16]  G. McLachlan.  The efficiency of Efron's “Bootstrap” Approach Applied to Error Rate Estimation in Discriminant Analysis , 1980 .

[17]  M. R. Mickey,et al.  Estimation of Error Rates in Discriminant Analysis , 1968 .

[18]  B. Efron.  Estimating the Error Rate of a Prediction Rule: Improvement on Cross-Validation , 1983 .

[19]  G. McLachlan.  AN ASYMPTOTIC EXPANSION OF THE EXPECTATION OF THE ESTIMATED ERROR RATE IN DISCRIMINANT ANALYSIS1 , 1973 .

[20]  G. McLachlan.  ASSESSING THE PERFORMANCE OF AN ALLOCATION RULE , 1986 .

[21]  William G. Cochran,et al.  Commentary on “Estimation of Error Rates in Discriminant Analysis” , 1968 .