Take this class if you enjoy watching paint dry. This guy is probably the WORST professor I have come across at Columbia. He has zero ability to teach and his lectures are completely and utterly inaudible. He would walk in and just start writing down illegible equations on the board with absolutely no context whatsoever and when students would ask clarifying questions he answered with something completely unrelated (or maybe I just didn't understand him because NO ONE CAN). He speaks at a volume of -10 and even if you could hear him it would be impossible to distinguish any words from his speech because he has such a strong accent. Sitting in his class was the most miserable, stressful, and tragic thing I have ever made myself do. On three occasions I simply walked out halfway through because I was getting so frustrated and angry at this joke of a class. I resorted to learning everything by myself from the textbook but even that was difficult and inefficient. I also just stopped going to class for a period of time because of the stress of not being able to hear a word he said. The TA's are not helpful either so recitations won't save you. There is absolutely no guidance in this course. Please please save yourself and do not take this class. You are paying MONEY to go to this school so you can have REAL TEACHERS! While Bai may be a smart guy he does not know how to teach and you will learn nothing from his dull painful lectures and extremely difficult exams. Absolutely not what I came to Columbia for...
Jushan's class is just okay. His workload is objectively alright compared to the complaints I've read about Seyhan's class. All his problem sets come straight from the textbook, so there aren't that many "custom" questions for you to wrangle with if you know what I mean. The class is boring. He just talks and writes on the chalkboard and pauses maybe once or twice per lesson to answer a question. Most people seem disinterested and only stay to copy his proofs which he tests in the Midterm and Final. I honestly learned more from studying the textbook than from attending his class. My main gripe with this class is that he provides no way to learn, other than to tell you to read the textbook. Unlike Seyhan, he does not provide any slides or learning material. You have to figure out R on your own. Last note: his class is objectively more theoretical than any other econometrics class. He is very fond of proofs and derivations, making this topic even more incomprehensible. The final exam barely had any questions that involved just plugging in numbers. In fact, almost all of them were either proofs he went through in class or him just testing you on a specific model. Get ready to barely learn the basics of a certain model and have him test you on a special variation of it in the exam. I wouldn't personally recommend this class unless you know what you're doing and you like learning on your own. People say he's nice, but what's the point if you dread going to class in the first place? My last gripe is that he also doesn't tell you what he's going to talk about in the next class, so you're left trying to interpret what he's saying as he tries to introduce (yet) another topic that you should've read up in advance.
One of the better Econometrics classes—the material isn't exciting (or at least it wasn't for me), but Bai is very clear in his lectures and it's easy to follow along with the textbook, which is surprisingly helpful. He's also really sweet in general. You're allowed to have a two-sided cheat sheet for the midterm, and two two-sided cheat sheets for the final; he really emphasizes what proofs and concepts you're supposed to know, so for some questions I ended up just copying over entire proofs from my sheet. The midterm average was 74.8/90, and the average for the final was 70.7/95. The pset questions are mostly from the textbook or proofs he's gone over in class/recitation (homeworks are graded from 0-100). The one issue I had with the class was that some homework questions required using R and even though the solutions were straightforward, looking up syntax took forever because we didn't go over R at all in class OR recitation. It wasn't a big deal but made doing homework a little more frustrating. Overall, I came in not remembering stats at all and still ended up with a good grade. The tests aren't crazy.
heard bai is teaching this again next year so wanted to post this (took it 2 years ago). if you want to actually learn metrics, this is the arguably the best intro to metrics columbia has to offer. personally i find metrics dry / wouldn't want to pursue this field further, but still did well in this class. he's good about reviewing stats before getting into the harder material. i also don't recall there ever being a mistake or typo on a pset or exam, and everything's pretty clear. overall a chill/nice prof, plus exams are fair, and he pretty much covers problems you'll see on the exams in class - so going to lecture is a good idea.
Jushan Bai's econometrics class is a solid class. Not mindblowingly amazing, but if I had to take econometrics again I would definitely take it with him. He's passionate about the topic but his lectures are still dry. If you're good at reading textbooks, then you'd be fine skipping, but all the material covered in homework and exams is covered in lecture so I think I saved a lot of time studying by attending most lectures. Prof Bai is also very helpful in office hours, so if you're having trouble or have any questions about the psets, he'll answer them for you. The workload was very reasonable and I didn't find it difficult despite not having a very solid foundation in statistics. He reviewed the topics we used (like hypothesis testing) at the beginning of the course, and I'd probably have been fine without having taken any statistics in the past. He provides wonderful pdf notes reviewing probability and statistics, which were super useful. The psets didn't involve that much R (I've heard horror stories about other classes and Stata) and were manageable. Overall, a medium level difficulty class with a pretty dedicated professor, and possibly one of the best metrics classes at Columbia.
The professor presents the material very clearly and with an excitement that makes one want to pay attention for a subject matter most consider boring. It's a very laid back class with the professor lecturing for the first month and students giving presentations for the rest of the semester. The proposal is due after the first few weeks, then two presentations spaced throughout the semester. You can get away without presenting until the last week, but be warned start on the paper early! Give yourself deadlines throughout the semester otherwise you'll come to reading week and still have 30 pages to write.