Necessary conditions for optimum truncated sequential tests. Simple hypotheses.

Autores: Enrique Castillo Ron , J. García
Localización: Stochastica: revista de matemática pura y aplicada, ISSN 0210-7821, Vol. 7, Nº. 1, 1983, págs. 63-81
Idioma: español
Títulos paralelos:
- Condiciones necesarias para pruebas secuenciales truncadas óptimas. Hipótesis simples.
Enlaces
- Texto completo (pdf)
Resumen
- The paper presents a new methodology to obtain partially sequential truncated tests which are optimum in the sense of minimizing the maximum expected sample number. This methodology is based on a variational approach and uses the Lagrange multipliers technique in order to obtain necessary conditions for a test to be optimum. By means of these conditions the optimum test can be obtained. Finally, the method is applied to the problem of testing the mean of an exponential distribution. As a particular test Wald's tests can be obtained in practice by using a sufficiently large truncation sample size.