Browsing by Author "Duarte Silva, A. P."
Now showing 1 - 10 of 23
Results Per Page
Sort Options
- Approaches to linear discriminant analysis of interval dataPublication . Duarte Silva, A. P.
- Asymptotic frameworks for high-dimensional two-group classificationPublication . Duarte Silva, A. P.Asymptotic properties of two-group supervised classi cation rules designed for problems with much more variables than observations are discussed. Two types of asymptotic bounds on expected error rates are considered: (i) bounds that assume consistent mean estimators and focus on the impact of the covariance matrix estimation. (ii) bounds that consider the errors in mean and covariance estimation. Known results for independence-based classi cation rules are generalized to correlationadjusted linear rules.
- Calibrating and estimating exponential-affine models in RPublication . Duarte Silva, A. P.; Cruz, Lia Vaz
- Classificação de dados de elevada dimensão: ignorar ou incorporar?Publication . Duarte Silva, A. P.
- Classificação supervisionada para dados de elevada dimensãoPublication . Duarte Silva, A. P.
- Classifying High-Dimensional Data with the The HiDimDA packagePublication . Duarte Silva, A. P.
- Comparing High-Dimensional Classifiers: Abuse and dangers of overall accuracyPublication . Duarte Silva, A. P.Statistical classification has a respected tradition in the support of medical diagnosis. Early applications relied on classical methodologies that assumed training samples with more patients than disease predictors and understood that simple performance measures, that do not take into account disease prevalence and the different costs of negative and positive predictions, have serious limitations. More recently, new classification methodologies have been applied to large genomic data bases where thousands of genes are measured on a few dozen patients. However, many of the studies that have evaluated these proposals employed only overall accuracy measures. This practice is potentially misleading, as it is known that changing prior probabilities and/or cost assumptions can strongly affect the relative standing of traditional classification rules. This presentation describes a study on the consequences of comparing high-dimensional classification rules by different performance measures. It will be argued that measures based on expected utilities or decision curves, that focus on the precision of risk estimates near the optimal threshold, should be preferred to overall accuracy. Furthermore, it will be shown that when samples proportions are not close to true disease probabilities corrected by misclassification costs, the use of overall accuracy can indeed lead to incorrect rankings of high-dimensional classifiers.
- Discriminant Analysis of Interval Data: Parametric Versus Distance-Based ApproachesPublication . Duarte Silva, A. P.; Brito, Paula
- «
- 1 (current)
- 2
- 3
- »