U of T Statistical Sciences: Congratulations on your Chapman fellowship. Can you tell me what a day in the life of a Chapman fellow is like?
Yanbo Tang: Well, it's mostly similar to what I would describe my PhD experience as, except more responsibility. Right now, for example, I'm teaching a class that’s taking a bit of time because of all the material I have to make for it and so on. I guess half the time I spent on my classes, the other half the time I would say I spend on research, and then a bit of admin tasks.
I've also just recently moved to the UK so it's been a bit of a battle to get everything settled.
Right! How's that? How are you enjoying London?
It's been nice so far! People say it's supposed to be raining all the time, but it's been only raining half the time, so I'm glad about that.
Can you tell me a bit about your background and how you got interested in statistics?
I guess I began what you can call my career back when I was in undergrad, where I studied actuarial mathematics, which is a kind of mathematics that’s meant to be applied to insurances and financial instruments and stuff like that.
I finished my degree and I worked for a bit, but I didn't quite like it. It wasn't really what I wanted to do. I actually enjoyed thinking about problems in my free time and stuff, and working wasn't quite providing the stimulation that keeps life interesting, I guess.
So that's why I moved on to doing a masters in statistics at the University of Toronto, just to try to see what I could do that's more critical and/or what could keep me thinking across the day about problems that I find interesting. And then I stayed at U of T to do my PhD because I figured that's what I liked to do.
What drew you to transition from actuarial science to a statistics program?
I guess in terms of the subject matter itself, it's because it's close to actuarial mathematics, in the sense that it's not too hard of a jump. If I went into something like pure mathematics, it'd be too hard, because I never had the formal training in that kind of thing.
Actuarial science is essentially a way of more or less taking probability into account when calculating streams of money. Like, if you think about retirement, for example, it's actuarial. You have to make sure you get enough money to invest into whatever assets you want in order to pay off people's pension rate. But at the same time, you want to make sure that people don't overpay you and you give too little, and vice versa.
So, I guess in that sense it's mostly dealing with randomness and seeing how that randomness impacts real life phenomena, which is what statistics does as well. In a broad sense, statistics looks to see how to account for randomness in experiments, for example, or in everyday life, in a way that you can still say something interesting or meaningful about what you study.
In terms of the program at U of T, overall I heard very good things about it and I met some of the profs who were teaching there and they all seemed quite nice and approachable. And then I figured it’d be good to stay in it for my PhD.
So what are you currently working on, research-wise?
Most of the time I've been focusing on my class. I’ve taught before, but it's a bit stressful to adapt to teaching at a new institution. Research wise, I'm working on projects on high dimensional statistics, which is a subfield of critical statistics. But I have been doing less work on that than I would've liked. So it's mostly like getting things ready slash working on just the class I'm teaching.
And what class is it?
Probability for statistics. It's a class on probability theory for the graduate students here.
Okay, cool. I saw that you won a presentation award at the SSC 2022 conference in the probability section.
Oh yeah, I did.
Can you talk a bit more about that experience and that project?
Yeah, sure. So to give context, the award is given in these presentation sections. Everyone who's presenting on a subject that's related to probability gets grouped together and you would kind of compete with each other. It’s more so that someone validates your talks based on certain criteria and they give like two or three prizes to people who they thought did a good job.
So the one I spoke about was dimensions, which is kind of like describing what happens if you let the problem you're studying become more complex in a certain way, and seeing how that impacts the overall quality of what you're getting out of the process you're studying.
Yeah, it was nice! I didn’t think I'd win because it wasn't a very technical talk. I'm glad to have been considered.
Do you have any long term research goals?
Generally speaking, I like just thinking about problems I like and seeing where that takes me. That’s maybe a bit too broad of an answer. But right now I’m interested in high dimensional statistics, which I mentioned a bit earlier.
I guess I like to think about problems that I find interesting, but also are somewhat useful to people. I like to do problems that are related to things that we see in science and that we see overall in the real world. So that, you know, it has a meaningful impact on things that happen in reality, I guess.
If you had any advice for students who are considering a PhD in statistics, but were not sure whether it's the right fit for them, what advice or what would you have them consider?
It's always hard to give general advice, because it depends on the individual, right? To an extent, I feel like anything that applies to everyone will be meaningless in terms of statement, you know what I mean? Because then that's just like platitude that you say to people.
But I think specifically it's good to try to figure out what you wanna do in terms of your end goal first and work backwards. And so for example, if you feel like, “Oh, I wanna do a PhD because I feel like that’ll give me a boost in my career and in data science”, whatever. Sure. That’s good. Then you kind of roll that and see what you have to do to have that end goal met. Right?
So research wise, you might not want to do something too critical because that would not be seen well by someone who wants someone who can just do the work in terms of applied statistics and give results and, you know, work in a real life environment. Whereas if you like the more critical stuff, it could be good to work backwards from that and see what the end goal would be like for you.
It really depends because statistics is such a large field that you can do a lot with it, which is good. And that freedom is a blessing, but it’s also what harms you because you can do so many things that you're not sure what to do. So having a general idea of what you would like to do based on your personality and your general vibe and then working backwards from there would be helpful.
In terms of the actual PhD, I think everyone's PhD is a bit different somehow. It's weird. I talk to my friends and we all have similar experiences to a certain degree, but we all have our own idiosyncrasies. I guess for me, the best advice I ever got is that sometimes succeeding just means failing at harder and harder things. If you never failed at something before, eventually you need to because it means you never pushed yourself to do what you could do, right? I guess a PhD, to a certain extent, is meant to be about pushing a field forward or pushing yourself forward to confront harder and harder problems.
So it's okay to fail, and it's to be expected that you fail things, but then it’s about how you deal with failure and how you keep on going even though you might not feel good about yourself. It’s that kind of mental fortitude that you need to keep on going.
So what do you do for fun outside of statistics?
I like reading. I like reading fiction mostly, but some nonfiction as well.
What’s on your shelf? Any recommendations?
Recently I’ve been reading Stefan Zweig. I like his work. His novels are good and also his memoir is excellent. If I recommend anyone, it’s him. Particularly The World of Yesterday. It's about his life pre-World War I to the middle of World War II, and how that affected him as a human being and how it felt to see Europe as a kind of center of cultural enlightenment become worse and worse and worse until the world that he knew is just gone.
Like, the Europe where people were kind of all friends between nations and moving freely across nations -- that kind of world is gone now for him. That's what he missed the most about it. It’s really nice. It's almost sad to read it because but you’re just seeing him describe his memories and how much he enjoyed having that kind of freedom and how much he loved his friends in France. Just seeing that all go away across his lifetime is really sad.
What else is on your shelf?
The Master and Margarita is pretty good. It’s kinda silly, but there’s definitely a really nice message throughout it, I find. Brothers Karamazov was pretty nice, I guess it’s one of my favorites. I always find the line “Without God, everything is permissible” super funny.
Why’s that?
When Ivan said that, he meant it in its opposite way obviously, but then the clergy took it a completely different way. I thought that was cool.
You know, it actually relates back to The Master and Margarita, where in the beginning they speak about Emmanuel Kant’s six proof existence of god. The first five proofs are logical statements. And then Emanuel Kant's proof of the existence of God is that like, he is the absolute moral authority. So therefore you need that in everything in order to even have this idea of order and what is correct and what is not correct, and that's why God exists. And so in the Brothers Karamazov, “Without god, everything is permissible” is, to me, kind of linked to that, where it's about how you need to have some kind of idea of what is correct. Yeah, I like the Russians.
.jpeg?auto=webp)
.jpeg?auto=webp)
/1x.jpeg?auto=webp)
/1x.jpeg?auto=webp)
/1x.jpeg?auto=webp)
/1x.jpeg?auto=webp)
/1x.jpeg?auto=webp)
/1x.jpeg?auto=webp)