What's A Good Stats/ML Book That I Can Use To Prep For Interviews?

P

potentialquant

Joined
3/17/20
Messages
11
Points
11
Soon to be PhD grad and will be interviewing for quant roles (researcher or developer). My background is in scientific computing, with some exposure to machine learning through coursework, side projects, and it was a small component of my research that didn't make it's way into my thesis.

I wouldn't say I have a great background in stats and ML. I've never taken a probability theory class in grad school, but some probability theory was covered in my ML class. Is graduate level probability theory knowledge required for interviews, or is an undergrad level sufficient?

The last time I took a pure stats class was back in undergrad, so I definitely need a refresher. I have a pretty strong background in numerical linear algebra, numerical methods, and differential equations (I have NO background in stochastic calculus), but it seems the quant interviews will focus more on the statistical math than the mathematical domains I enumerated.

What would be a good book for me to use to prepare for stats and ML-styled questions?

I've been recommended the following:

Joshi - Quant interview and answer guide
Vault quant interview
Heard on the Street

but I believe all 3 of these are only useful for brainteasers and not for the theoretical stats/ML questions that I may be asked.

Thanks!
 
The undergraduate version (Introduction to Statistical Learning) is pretty good too. But if you're aiming for top quant funds that recruit PhDs, they will dig in and that will not be enough. Otherwise, it's solid.
 
quantsmodelsbottles said:
The undergraduate version (Introduction to Statistical Learning) is pretty good too. But if you're aiming for top quant funds that recruit PhDs, they will dig in and that will not be enough. Otherwise, it's solid.
Click to expand...
Wait, which book are you saying is not enough? EEP or Intro to SL? I assume you're referring to Intro to SL, but EEP looks more like a cursory overview rather than an in-depth guide?
 
quantsmodelsbottles said:
Yeah I am referring to Intro to SL. The first one is definitely not just a cursory overview.
Click to expand...
Ah okay thanks. I haven't read the EEP book at all; I just flipped through the pages and saw that it covered a lot of breadth in about 700 pages, so I just assumed that it's not super in-depth.
 
IMHO it's a good strategy to read through Elements of statistical learning. It's for someone who already has enough background in statistics and want to grasp certain topics/techniques. It's like a reference book, you go straight to that chapter which you are interested in.

potentialquant said:
Ah okay thanks. I haven't read the EEP book at all; I just flipped through the pages and saw that it covered a lot of breadth in about 700 pages, so I just assumed that it's not super in-depth.
Click to expand...
 
Lynette Zhang said:
I personally don't think it's a good strategy to read through EEP. It's for someone who already has enough background in statistics and want to grasp certain topics/techniques. It's like a reference book, you go straight to that chapter which you are interested in.
Click to expand...
I'm not sure if I fall into that "someone" category. The last time I took a pure stats course was in undergrad (5+ years ago). But I have taken 2 recent grad courses (ML and AI theory) that used probability quite a bit. Just by looking through the contents of EEP, it looks like most of the material was covered in my ML class. Outside of regression, deep learning, and occasionally optimization, I don't use much statistics/probability in my research, so most of my background in stats/probability have been acquired through coursework.

I'm trying to get a refresher for stats and probability as my current priority. The EEP book seems too ML focused, and I don't think it'd give me the stats/probability refresher that I feel I need. It seems like the stats/probability stuff asked during interviews are at the high school/college level.
 
Sounds like you're looking for more fundamental suggestions like bayesian statistics/inference and probability theory? There's tons of books on probability, but intro to probability by blitzstein is solid. He is a harvard teacher and his video lectures are online as well if you come across difficulty in the book. I used intro to probability models by sheldon ross in my undergrad; the book breaks down probability theory, but I think blitzstein will give you a stronger intuition. He has suggested problems on his site with solutions. Some are a bit tricky. Lots of good statistics books are written as econometrics books, but I haven't gotten there yet. I would suggest doing probability theory first because probability is the foundation for all of statistics and can close any small gaps you may have in stat.
 
princeofbelair said:
Sounds like you're looking for more fundamental suggestions like bayesian statistics/inference and probability theory? There's tons of books on probability, but intro to probability by blitzstein is solid. He is a harvard teacher and his video lectures are online as well if you come across difficulty in the book. I used intro to probability models by sheldon ross in my undergrad; the book breaks down probability theory, but I think blitzstein will give you a stronger intuition. He has suggested problems on his site with solutions. Some are a bit tricky. Lots of good statistics books are written as econometrics books, but I haven't gotten there yet. I would suggest doing probability theory first because probability is the foundation for all of statistics and can close any small gaps you may have in stat.
Click to expand...
Yeah basically. Although, I'm looking for more of a quick refresher than going through a textbook. I'm currently going through "Heard On the Street," and I'm still good on most of the stats concepts used in this book (mostly expected value, conditional probability, bayes theorem, just very basic stuff so far), so maybe that gives an idea of the level I'm at. I plan on also going through the "A practical guide to quantitative finance interview" book as well. I don't think either of these books cover things like "null hypothesis, p value, t value, regression, etc..," which I heard are occasionally asked at some of the firms (I'm not interviewing with banks, it's mostly hedge funds and prop trading firms).
 
Take a look into Probability and Statistics by DeGroot and Schervish. You'll find a pdf of it with a quick google search. It will have what your looking for in both a refresher for probability and statistical inference. Another good probability book that I read in undergrad was introduction to probability theory by Port Hoel and Stone. This book is older, but is pretty good. If your looking for a quick refresher though, I would go with DeGroot and Schervish.
 
YankeesR said:
Take a look into Probability and Statistics by DeGroot and Schervish. You'll find a pdf of it with a quick google search. It will have what your looking for in both a refresher for probability and statistical inference. Another good probability book that I read in undergrad was introduction to probability theory by Port Hoel and Stone. This book is older, but is pretty good. If your looking for a quick refresher though, I would go with DeGroot and Schervish.
Click to expand...
Thanks. I found the book. It's about 900 pages. Would you say all the topics in here are fair game for quant interviews? Just from looking at the TOC, it looks like most of it is relevant.
 
I would not say I'm qualified to answer what is needed for quant interviews and don't want to lead you in the wrong direction, other people on this forum surely can though. I just saw the topics you listed and from reading those books, they are definitely relevant to those topics, and good books I would add.
 
Back
Top Bottom