Comment on "On some statistical paradoxes and non-conglomerability" by Bruce Hill.
Levi, Isaac
Trabajos de Estadística e Investigación Operativa, Tome 32 (1981), p. 135-141 / Harvested from Biblioteca Digital de Matemáticas
Access to full text

Those who follow Harold Jeffreys in using improper priors together with likelihoods to determine posteriors have thought of the improper measures as probability measures of a deviant sort. This is a mistake. Probability measures are finite measures. Improper distributions generate σ-finite measures. (...)

Publié le : 1981-01-01
DMLE-ID : 3255 
@article{urn:eudml:doc:40657,
     title = {Comment on "On some statistical paradoxes and non-conglomerability" by Bruce Hill.},
     journal = {Trabajos de Estad\'\i stica e Investigaci\'on Operativa},
     volume = {32},
     year = {1981},
     pages = {135-141},
     zbl = {0506.62001},
     mrnumber = {MR0697204},
     language = {en},
     url = {http://dml.mathdoc.fr/item/urn:eudml:doc:40657}
}

Levi, Isaac. Comment on "On some statistical paradoxes and non-conglomerability" by Bruce Hill.. Trabajos de Estadística e Investigación Operativa, Tome 32 (1981) pp. 135-141. http://gdmltest.u-ga.fr/item/urn:eudml:doc:40657/