A Comparative Study of Kernel-Based Density Estimates for Categorical Data

Kernel estimates of discrete probabilities are considered, with emphasis on computation of the smoothing parameters. Different approaches based on minimum mean squared error, cross-validation and pseudo-Bayesian techniques are compared, particularly from the points of view of reliability and ease of computation. The advantages of a fractional allocation procedure and of computing the bandwidths marginally for each variable are pointed out. Multicategory variables and incomplete data can be coped with. The relationship between the kernel method and other smoothing techniques for categorical data is discussed.

[1]  J A Anderson,et al.  A statistical aid to the diagnosis of keratoconjunctivitis sicca. , 1972, The Quarterly journal of medicine.

[2]  J. Habbema,et al.  Selection of Variables in Discriminant Analysis by F-statistic and Error Rate , 1977 .

[3]  P. Holland,et al.  Simultaneous Estimation of Multinomial Cell Probabilities , 1973 .

[4]  M. Stone.  Cross-validation and multinomial prediction , 1974 .

[5]  Donald B. Rubin,et al.  Max-imum Likelihood from Incomplete Data , 1972 .

[6]  D. Titterington,et al.  Analysis of incomplete multivariate binary data by the kernel method , 1977 .

[7]  Colin Aitken,et al.  An Application of Discrete Kernel Methods to Forensic Odontology , 1979 .

[8]  B. Silverman,et al.  Choosing the window width when estimating a density , 1978 .

[9]  D. M. Titterington,et al.  Updating a Diagnostic System using Unconfirmed Cases , 1976 .

[10]  R. Fletcher.  A General Quadratic Programming Algorithm , 1971 .

[11]  A. F. Smith,et al.  A Quasi‐Bayes Sequential Procedure for Mixtures , 1978 .

[12]  R. Kronmal,et al.  Some Classification Procedures for Multivariate Binary Data Using Orthogonal Functions , 1976 .

[13]  D. Rubin,et al.  Maximum likelihood from incomplete data via the EM - algorithm plus discussions on the paper , 1977 .

[14]  A. F. M. Smith,et al.  A quasi-Bayes unsupervised learning procedure for priors (Corresp.) , 1977, IEEE Trans. Inf. Theory.

[15]  J. Aitchison,et al.  Multivariate binary discrimination by the kernel method , 1976 .

[16]  M. Stone.  An Asymptotic Equivalence of Choice of Model by Cross‐Validation and Akaike's Criterion , 1977 .

[17]  Robert P. W. Duin,et al.  On the Choice of Smoothing Parameters for Parzen Estimators of Probability Density Functions , 1976, IEEE Transactions on Computers.

[18]  Tom Leonard,et al.  A Bayesian Approach to Some Multinomial Estimation and Pretesting Problems , 1977 .

[19]  D. W. Roncek,et al.  Discrete Discriminant Analysis. , 1979 .

[20]  D. Titterington,et al.  Estimation Problems with Data from a Mixture , 1978 .

[21]  Tom Leonard,et al.  A Bayesian method for histograms , 1973 .

[22]  D. Moore.  Evaluation of Five Discrimination Procedures for Binary Variables , 1973 .