Characterizing Density Crossing Points.
The writers discuss where and how often distinct probability density functions cross. Such crossing points are of particular interest if, for example, the normal is deployed as an approximation according to the central limit theorem, as they determine the total variation distance and the areas of o...
| Publicado en: | American Statistician Vol. 61; no. 1; pp. 28 - 34 |
|---|---|
| Autores principales: | , , |
| Formato: | Artículo |
| Publicado: |
American Statistical Association
February 2007
|
| Materias: | |
| Acceso en línea: | Ver este registro en EBSCOhost |
| Sumario: | The writers discuss where and how often distinct probability density functions cross. Such crossing points are of particular interest if, for example, the normal is deployed as an approximation according to the central limit theorem, as they determine the total variation distance and the areas of over- and underestimation of actual probabilities. Furthermore, crossing points are strongly associated with the behavior of the density ratio, a ubiquitous quantity in many fields of statistics. Several examples intended to characterize density crossing points are provided. |
|---|