Add to ORKGA comparison of the error probabilities for various discriminating rules is performed in the two population cases when nothing is known of the populations other than they are bivariate negative exponential. In most cases, the absolute difference between the error probabilities for each function was very small. However, the Euclidean distance function consistently performed as well as, and sometimes superior to any of the others studied in the this thesis.http://archive.org/details/acomparisonofvar1094556193Lieutenant (junior grade), United States NavyApproved for public release; distribution is unlimited
A new methodology, based on the asymptotic separation of probability laws, was introduced by Gonçalv...
peer reviewedMonte Carlo experiments are performed to compare twenty estimators or error rates in di...
peer reviewedMonte Carlo experiments are performed to compare twenty estimators or error rates in di...
The performance of four discriminant analysis procedures for the classification of observations from...
An asymptotically non-parametric method was used to test the equivalence of the marginal distributio...
Various parametric and nonparametric approaches to multiple discriminant analysis attempt to discrim...
An asymptotically non-parametric method was used to test the equivalence of the marginal distributio...
peer reviewedA simulation study has been used to evaluate the minimal error rate of three affectatio...
Various parametric and nonparametric approaches to multiple discriminant analysis attempt to discrim...
Various parametric and nonparametric approaches to multiple discriminant analysis attempt to discrim...
AbstractThis article presents simulation results comparing various resampling estimators of classifi...
Building on probabilistic models for interval-valued variables, parametric classification rules, bas...
This study examined the error of classification associated with the gamma distribution for the linea...
The problem of discriminating between two n-variate normal populations with known but unequal means ...
When outcome is ordinal, it would be expected that classification procedures which assume ordering w...
A new methodology, based on the asymptotic separation of probability laws, was introduced by Gonçalv...
peer reviewedMonte Carlo experiments are performed to compare twenty estimators or error rates in di...
peer reviewedMonte Carlo experiments are performed to compare twenty estimators or error rates in di...
The performance of four discriminant analysis procedures for the classification of observations from...
An asymptotically non-parametric method was used to test the equivalence of the marginal distributio...
Various parametric and nonparametric approaches to multiple discriminant analysis attempt to discrim...
An asymptotically non-parametric method was used to test the equivalence of the marginal distributio...
peer reviewedA simulation study has been used to evaluate the minimal error rate of three affectatio...
Various parametric and nonparametric approaches to multiple discriminant analysis attempt to discrim...
Various parametric and nonparametric approaches to multiple discriminant analysis attempt to discrim...
AbstractThis article presents simulation results comparing various resampling estimators of classifi...
Building on probabilistic models for interval-valued variables, parametric classification rules, bas...
This study examined the error of classification associated with the gamma distribution for the linea...
The problem of discriminating between two n-variate normal populations with known but unequal means ...
When outcome is ordinal, it would be expected that classification procedures which assume ordering w...
A new methodology, based on the asymptotic separation of probability laws, was introduced by Gonçalv...
peer reviewedMonte Carlo experiments are performed to compare twenty estimators or error rates in di...
peer reviewedMonte Carlo experiments are performed to compare twenty estimators or error rates in di...