Review Comment

W4150 Probability and Statistics

May 26, 2014

Rabinowitz, Daniel
W4150 Probability and Statistics

Please keep in mind that this review is more than 5 years old.

A rare and exceptional professor.

I have never seen a professor work so hard and care so much for his students. He provided every imaginable resource, including optional weekly session to learn SAS and he wrote up complete packets detailing exactly what we would need to know for each exam. He was always quick to respond to email and in the online class discussion with a detailed answer. I am reluctant to write here the extent to which he modified the course to accommodate the requests of students, because it would be a shame to see people taking advantage of this guy. But I will say that everything but a short quiz and 2 exams is optional, and he does everything imaginable to make sure you get a good grade if you want it.

The structure of the course treats the student with a lot of respect - you have all the resources you could want to learn as much as you like, but in terms of the grade you are only accountable for understanding the material on the exams. The downside to this is that when you are overwhelmed with the daily nonsense of other courses (I'm looking at you Core) , it is easy let all the great and important optional stuff slide. Even though its optional, it would behoove you to at least try the homework. Also, a lot of the final exam preparation boiled down to figuring out what sly little algebra trick was used in a step of a proof: This is a curious exercise, but it became a viscous time suck that had little to do with the substance of the proof, its meaning or application. Not difficult, just tedious. I gather it was the first time he attempted this format, so perhaps it will be less of a problem for the next batch.

Rabinowitz doesn't merely lecture, he communicates. In my experience here, this is unusual for a math or science professor. As a math major, I've had many brilliant professors, but most of them simply recite the material, and make the students into passive observers taking dictation. Rabinowitz would regularly check in with us, engage us in little experiments, make up interesting stories and examples to summon our attention in the dry spots. Most importantly, he was always sensitive to when something might not be clear, and always ready to patiently explore different ways of trying to get the idea across. He seemed personally devastated when he saw people uninterested or not following, and I found this inspired in me to be less apathetic and stupid.

The material itself is important and fascinating (I think). The course is theory heavy, but he spent a good amount of time on examples and more practical matters. His exams were almost entirely theory, but it is beautiful theory. I got the impression that he was trying to drive it more towards applications in the final weeks, and might place more emphasis on applications in future iterations of the course. I've heard nightmares about the stat department, but this course was a gem, and I'm now doing more probability theory and statistics specifically because of my experience here.

So- great guy, brilliant teacher, interesting and useful course.

Workload:

Depends on you. required for the grade: a small quiz, midterm and final (55%), all open notes. All multiple choice, no partial credit (at least on final). Complete and (almost) exhaustive review material distributed before each exam. Except for some very general and easy applied problems, the exam is taken directly from the review material, which you can mark up and bring with you. He seemed to be always tweaking the course, so it might be different for you. Lots of optional stuff (homework, SAS) to gain higher degrees of mastery.

December 21, 2010

Gallego, Guillermo
W4150 Probability and Statistics

Please keep in mind that this review is more than 5 years old.

Gallego gets a bad rap. This is a course in theoretical mathematics, not AP statistics, and he makes that perfectly clear on the first day---if you don't want rigor, take the 3000-level version of this class. Lectures are about theorems and their proofs, and that's the way it should be. Gallego himself is clear, concise, gives reasonable exams (too easy, if anything; a 98 on the midterm was an A-) and assigns (mostly) useful homework. The only complaints I have are that his lecture notes are absolutely unreadable (he writes them down on his computer, and his handwriting is atrocious), and that more than the one or two theoretical homework problems he assigns (rather than number-crunching problems from the book) would be useful.

The students are the real problem in this class. Evidently it's a requirement for OR, FE and econ, so it's made up mostly of students who obsess constantly about their grades but who are breathtakingly uninterested in the material (or, for that matter, in anything other than being rich). Highlights included gasps of outrage at the aforementioned midterm curve, and a large puddle of saliva on the floor after a sample problem involving a financial instrument.

Workload:

10 psets -- I started most of them the night before they were due and had no trouble.
1 midterm, 1 final -- if you do better on the final, he drops your midterm completely

No psets dropped, but a generous helping of late days.

June 14, 2005

Gallego, Guillermo
W4150 Probability and Statistics

Please keep in mind that this review is more than 5 years old.

Gallego assumes that you know a lot more than you do to begin wtih. Of course, this is an advanced class and the students know basic statistical methods. However, he presents his information in proofs that are confusing and difficult to understand. However, if I look the same information up in the textbook or elsewhere, I realize that what Gallego was trying to teach is something relatively simple, or at least understandable when presented more simply. Basically, he unecessarily complicates theories and equations. The textbook was excellent, however, so that helped when I had to teach myself material. The fact that he did not prepare us well was apparent when most of the students scored between about 20-30 points on the 100 point midterm; he was forced to give a makeup midterm.

Workload:

weekly problem sets, usually not too taxing; very difficult midterm and final

Directory Data

Dept/Subj Directory Course Professor Year Semester Time Section