American Journal of Theoretical and Applied Statistics
Volume 8, Issue 6, November 2019, Pages: 203-213
Received: Mar. 2, 2018;
Accepted: Mar. 30, 2018;
Published: Nov. 8, 2019
Views 94 Downloads 48
Awogbemi Clement Adeyeye, Statistics Department, Nnamdi Azikiwe University, Awka, Nigeria
Onyeagu Sidney Iheanyi, Statistics Department, Nnamdi Azikiwe University, Awka, Nigeria
This study investigates the errors of misclassification associated with Edgeworth Series Distribution (ESD) with a view to assessing the effects of sampling from non-normality. The effects of applying a normal classificatory rule when it is actually a persistent non-normal distribution were examined. These were achieved by comparing the errors of misclassification for ESD with ND using small sample sizes at every level of skewness factor. The simulation procedure for the experiment of the study was implemented using numerical inverse interpolation method in R program to generate a uniformly distributed random variable N. A configuration size of 1000 was obtained for the two training samples drawn at every level of skewness factor (λ3), in the range (0.00625, 0.4). This was repeated for different small sample sizes by comparing errors of misclassification of ESD with ND. The simulation results showed that the optimum probabilities of misclassification by ESD: (E12E) decreases and (E12E) increases, as the skewness factor (λ3) increases. The optimum total probability of misclassification is stable as λ3 also increases. The probability of misclassification E12E ≥E12N and E21E ≥E21N at every level of λ3. Thus, the total probabilities of misclassification are not greatly affected by the skewness factor. This asserts that the normal classification procedure is robust against departure from normality.
Awogbemi Clement Adeyeye,
Onyeagu Sidney Iheanyi,
Errors of Misclassification Associated with Edgeworth Series Distribution (ESD), American Journal of Theoretical and Applied Statistics.
Vol. 8, No. 6,
2019, pp. 203-213.
Alvin, C. R. (2002). Methods of Multivariate Analysis, 2nd ed. John Wiley & Sons, New York, pp. 270.
Anderson, T. W. (2003). An Introduction to Multivariate Statistical Methods, New York: John Wiley, pp. 126-127.
Barton, D. E. and Dennis, K. E. (1952). The Condition under which Gram-Charlier and Edgeworth Curves are Positive Definite and Unimodal, Biometrika, 39, pp. 425- 427.
Bhuyan, C. K. (2010). Probability Theory and Statistical Inference, 1st ed., New Central Book Agency (Ltd), London, pp. 203.
Broffit, J. D. and Williams, J. S. (1973). Minimum Variance Estimators for Misclassification Probabilities in Discriminant Analysis, Journal of Multivariate Analysis, (3), pp. 311-327.
Chingada, E. F. and Kocherlakota, S. (1978). Robustness of the Linear Discriminant Function to nonnormality, Journal of Statistical planning and Inference, 3, pp. 74.
Draper, N, and Tierny, D. (1972). Region of Positive and Unimodal Series Expansion of the Edgeworth and Gramcharlier Approximations. Biometrika, 59 (2), pp. 463-465.
Fisher, F. A. (1936). The Use of Multiple Measurements in Taxonomic Problems, Annals of Eugenics, 7, pp. 179-188.
Geisser, S. (1967). Estimation Associated with Linear Discriminants. Annals of Mathematical Statistics, (38) pp. 807-817.
John, P. B (2010). Measurement Error, Model, Methods and Applications, CRC Press, Taylor and Francis, pp. 1-10.
Kendall, M. G. and Stuart, A. (1958). The Advanced Theory of Statistics, Vol. 3, Hafner: New York. pp. 155.
Kocherlakota, S., Kocherlakota, K. and Balakrishnan, N (1987). Expansions of Errors of Misclassification, Advances in Multivariate Statistical Analysis, pp. 191-211.
Lachenbruch, P. A., Clarke, B. B. & Lin, L. (1977). The Effect of Non-normality on Quadratic Discriminant Function, MEDINFO 77 Shives/Wolf IFEP, North Holland Publishing Co. pp. 101-104.
Lachenbruch, P. A., Sneeringer, C. &Revo, L. T. (1973). Robustness of the Linear and Quadratic Discriminant Function to Certain Types of Non-normality. Journal of Communication Statistics, 1, pp. 39–57.
Mahmoud, M. A. and Moustafa, H. M. (1995). Errors of Misclassification Associated with Gamma Distribution, Journal of Mathematical Computing Modeling, Vol. 22, No. 3, pp. 105-119.
Ogum, G. E. O. (2002). Introduction to Methods of Multivariate Analysis, Afri-Towers Ltd., Aba, Nigeria. pp. 107.
Peter, H. (1997). The Bootstrap and Edgeworth Expansions, Springer Series in Statistics, pp. 145.
Richard, A. J. and Dean, W. W. (1988). Applied Multivariate Statistical analysis, 3rd ed., Prentice Hall, Inc. New Jersey, pp. 497.
Ruby, C. W. (2010). A Bayesian Edgeworth Expansion by Stein’s Identity, Bayesian Analysis, 5 (4), pp. 741-764.
Sedransk, N. and Okamato, M. C. (1971). Estimation of the Probabilities of Misclassification for the Linear Discriminant Function in the Univariate Normal Case. Annals of Statistics, 23 (3), pp. 419-427.
Toussaint, G. T. (1974). Bibliography on Estimation of Misclassification. IEEE Transactions on Information Theory. Vol. 2 (4), pp. 472–479.
William, R. D. and Matthew, G. (1984). Multivariate Analysis, Methods and Applications. John Wiley & Sons, Inc. N. Y. pp. 360-361.