In considering problems regarding the foundations of probability, I have
thought that Brouwer's choice sequences could be an interesting basis
for measure theory and probability theory. The class of choice sequences
can be regarded both as the codomain and (in essence) as the domain of
probability functions. A such approach could suggest possible solutions
to some classical (and nonstandard) measure and probability foundational
problems, such as that of the mostly forcedly "dogmatic" (or
"definitively" infinitesimal) prior assignments in a probability space.
Why don't develop a measure theory and a probability theory based on
choice sequences? Why not?
Giovanni Lagnese
__________ Informazioni da ESET NOD32 Antivirus, versione del database delle firme digitali 5387 (20100823) __________
Il messaggio è stato controllato da ESET NOD32 Antivirus.
www.nod32.it