Single-Step Simple ROC Curve Fitting via PCA.
A simple approach to fitting curves to receiver operating characteristic rating data is presented. It is based on the first principal component of the covariance space of the inverse normal integral of the cumulative rating data of the targets and distractors. It provides for 2 new associated d' est...
| Publicado en: | Canadian Journal of Experimental Psychology / Revue Canadienne de Psychologie Expérimentale Vol. 70; no. 4; pp. 301 - 306 |
|---|---|
| Autor principal: | |
| Formato: | Artículo |
| Publicado: |
Canadian Psychological Association
Dec2016
|
| Materias: | |
| Acceso en línea: | Ver este registro en EBSCOhost |
| Sumario: | A simple approach to fitting curves to receiver operating characteristic rating data is presented. It is based on the first principal component of the covariance space of the inverse normal integral of the cumulative rating data of the targets and distractors. It provides for 2 new associated d' estimates, d' and d'. A Monte Carlo simulation demonstrated that the parameter estimates are unbiased and produce estimates comparable to the iterative, maximum likelihood approach. The corresponding computational and plotting functions in the R programming language are also provided. |
|---|