Statistics and Measure Theory

[qqyyzz]

I was reading an article on statistics education and there was a quote by Persi Diaconis about how there's a trend in statistics departments to no longer include measure theory as a requirement.  Every department that I've looked at requires measure theory (with the exception of UC berkeley).  I was wondering if you folks would know which departments no longer require measure theory for their PhD programs.

[cyberwulf]

I suspect that the trend is towards "overlooking" lack of a previous measure theory course for otherwise promising applicants. Many departments which are behaving this way may still have it officially listed as a prerequisite, however.

[timmmythetooth]

cyberwulf- I'm not sure I saw a single statistics department that had measure theory as a prerequisite when I was applying.  Maybe one or two that said it could be helpful.

 

Regarding the original question, a Harvard PhD stats grad I talked to said that the department there teaches very little measure theory.

[qqyyzz]

Cyberwulf, when I was applying last year not a single program had measure theory as a prereq for starting a program.  I'm 100% sure Persi Diaconis is talking about "modern" departments do not require students to take measure theory.  From reading UC Berkeley's program description it appears 215 (the measure theoretic probability sequence) is optional; there are other classes you can take as part of core.  I think UCLA's program is the same way.  Anyone know of any other programs like this?  I'm curious which other programs are "modern".

[cyberwulf]

I was taking OP at their word regarding departments listing measure theory as a pre-requisite. I'm not at all surprised to hear that most actually do not require it.

[DMX]

Cyberwulf, when I was applying last year not a single program had measure theory as a prereq for starting a program.  I'm 100% sure Persi Diaconis is talking about "modern" departments do not require students to take measure theory.  From reading UC Berkeley's program description it appears 215 (the measure theoretic probability sequence) is optional; there are other classes you can take as part of core.  I think UCLA's program is the same way.  Anyone know of any other programs like this?  I'm curious which other programs are "modern".

 

From anecdotal evidence from Faculty/current PhD students, it seems like fewer and fewer stat programs require it (as part of their PhD curriculum).

 

I would guess that this is in part due to Stats becoming more computational in nature.

[qqyyzz]

From anecdotal evidence from Faculty/current PhD students, it seems like fewer and fewer stat programs require it (as part of their PhD curriculum).

 

I would guess that this is in part due to Stats becoming more computational in nature.

Right, a ton of people are saying fewer and fewer programs require it, but I have literally only seen like two or three programs where it isn't required.  Where are these other programs that are causing people to say that fewer programs require it?

[Y.T. Safire]

Cyberwulf, when I was applying last year not a single program had measure theory as a prereq for starting a program.  I'm 100% sure Persi Diaconis is talking about "modern" departments do not require students to take measure theory.  From reading UC Berkeley's program description it appears 215 (the measure theoretic probability sequence) is optional; there are other classes you can take as part of core.  I think UCLA's program is the same way.  Anyone know of any other programs like this?  I'm curious which other programs are "modern".

 

The Columbia biostats department just added the course of measure-theoretic probability theory in the math department to the compulsory curriculum of its PhD students and the reason is to make it more "modern". It seems to be moving in the opposite direction under the same "name".