Three classifiers, namely linear discriminant analysis (LDA), quadratic discriminant analysis (QDA) and regularized discriminant analysis (RDA) are considered in this study for classification based on MR data. Because NIR data sets are severely ill-conditioned, the three methods cannot be directly a
Discriminant analysis of survey data
β Scribed by Ching-Ho Leu; Kam-Wah Tsui
- Publisher
- Elsevier Science
- Year
- 1997
- Tongue
- English
- Weight
- 873 KB
- Volume
- 60
- Category
- Article
- ISSN
- 0378-3758
No coin nor oath required. For personal study only.
β¦ Synopsis
We consider the problem of the effect of sample designs on discriminant analysis. The selection of the learning sample is assumed to depend on the population values of auxiliary variables. Under a superpopulation model with a multivariate normal distribution, unbiasedness and consistency are examined for the conventional estimators (derived under the assumptions of simple random sampling), maximum likelihood estimators, probability-weighted estimators and conditionally unbiased estimators of parameters. Four corresponding sampled linear discriminant functions are examined. The rates of misclassification of these four discriminant functions and the effect of sample design on these four rates of misclassification are discussed. The performances of these four discriminant functions are assessed in a simulation study.
π SIMILAR VOLUMES