Page 270 - Introduction to Statistical Pattern Recognition

252 Introduction to Statistical Pattern Recognition

10. Let the bootstrap sample covariance matrix be
..N ** r*
Z = Ewj(Xj - M )(Xj - M )T
j=l
instead of (5.172). Compute the expected value of i.

References

1. G. J. McLachlan, Error rate estimation in discriminant analysis: recent
advances, in “Advances in Multivariate Statistical Analysis,” ed. A. K.
Gupta, D. Reidel Publishing Co., Dordrecht, 1987.
2. D. J. Hand, Recent advances in error rate estimation, Patrern Recognition
Letters, 4, pp. 335-346, 1986.
3. A. K. Jain and B. Chandrasekaran, Dimensionality and sample size con-
siderations in pattern recognition practice, in “Handbook of Statistics
2,” ed. P. R. Krishnaiah and L. N. Kanal, North Holland, Amsterdam,
1982.
4. K. Fukunaga and R. R. Hayes, Effects of sample size in classifier design,
Trans. IEEE Pattern Anal. and Machine Intell., PAMI-1 1, pp. 873-885,
1989.
5. G. F. Hughes, On the mean accuracy of statistical pattern recognizers,
Trans. IEEE Inform. Theory, IT-14, pp. 55-63, 1968.
6. K. Fukunaga and R. R. Hayes, Estimation of classifier performance,
Trans. IEEE Pattern Anal. and Machine Intell., PAMI-11, pp. 1087-
1101, 1989.

7. S. Raudys and V. Pikelis, On dimensionality, sample size, classification
error, and complexity of classification algorithm in pattern recognition,
Trans. IEEE Pattern Anal. Machine Intell., PAMI-2, pp. 242-252, 1980.
8. D. H. Foley, Considerations of sample and feature size, Trans. IEEE
Inform. Theory, IT-18, pp. 618-626, 1972.
9. S. John, Errors in discrimination, Ann. Math. Stat., 32, pp. 1125-1144,
1961.
10. C. P. Han, Distribution of discriminant function in circular models, Inst.
Srat. Math. Ann., 22, pp. 1 17-125, 1970.