Le lundi 23 mai 2011 à 11:38 -0500, Matthew Rocklin a écrit : > Hi Everyone, > > > I'm starting work now on my GSoC project on random variables. I'll be > writing up technical plans on this wiki page > https://github.com/sympy/sympy/wiki/Random-Variables > It'll be updated as I go. > Please let me know if you have comments / see an issue / want to > suggest something. > Sets will be useful in many areas, not just your project, so I think you should start with that.
I think you should make your definition more rigorous, by giving symbolic names to things, to make it clearer what information each object requires. Your definition of probability spaces is a bit confusing. If you're using the measure-theoretic definition (and you basically have to, in order to handle continuous distributions correctly), you should explain why the measurable sets need to be specified. Anyway, an event is just any subset of the sample space that has a well-defined probability. For random variables, there's a problem: they're mathematically defined as functions from a hidden probability space into an observable sample space. The hidden space is required in order to handle correlations, but your definition doesn't seem to have an equivalent of that. Finally, the syntax 'Practice = Rain==False or Temp > 20' cannot work: neither '==' nor 'or' can be altered from their builtin meaning. You should use something like 'practice = ~rain | (temp > 20)'. But then the concept of event seems redundant with that of random boolean (if rain is a random boolean, certainly ~rain is one too). -- You received this message because you are subscribed to the Google Groups "sympy" group. To post to this group, send email to sympy@googlegroups.com. To unsubscribe from this group, send email to sympy+unsubscr...@googlegroups.com. For more options, visit this group at http://groups.google.com/group/sympy?hl=en.
