1–10 of 21 posts
What is your favorite book that transformed the way you interpret and produce statistics?
Anyway, about the books...
Blitzstein's Introduction to Probability and Harvard's Stat110 course are a good starting point if you've taken calculus. There's also good books like All of Statistics by Wasserman and Bayesian Data Analysis by Gelman but they're an absolute slog to get through - definitely not my favorite in any way but they cover a lot of stuff and you can have a copy around for reference.
intro: freedman, Pisani, purves. Very clear and accessible.
Intermediate/advanced: casella and Berger
Advanced: Bickel and doksum
Overview of ML and modern stat methods: efron and hastie
You are spoiled for good choices frankly.
What’s the best textbook you have read about X? In general the answer is “the third one.” By that time things sink in and the third book seems super clear and understandable.
https://www.goodreads.com/book/show/43722897-the-art-of-stat...
At what level? intro: freedman, Pisani, purves. Very clear and accessible. Intermediate/advanced: casella and Berger Advanced: Bickel and doksum Overview of ML and modern stat methods: efron and hastie You are spoiled for good choices frankly. What’s the best textbook you have read about X? In general the answer is “the third one.” By that time things sink in and the third book seems super clear and understandable.
This book explains the concepts without using mathematics, but even people with phDs in mathematics praise it as one of the best textbooks on statistics [1]
I really enjoyed https://xcelab.net/rm/statistical-rethinking/
It is very clearly written, full of practical examples (with code), and doesn't assume heavy math knowledge.
This section from the preface sums up the intended audience:
"The principle audience is researchers in the natural and social sciences, whether new PhD students or seasoned professionals, who have had a basic course on regression but nevertheless remain uneasy about statistical modeling. This audience accepts that there is something vaguely wrong about typical statistical practice in the early twenty-first century, dominated as it is by p-values and a confusing menagerie of testing procedures. They see alternative methods in journals and books. But these people are not sure where to go to learn about these methods."
It’s also not aimed at complete beginners, but if you have a very solid grasp on high school math and some blood on your teeth, you can totally do it.
There’s also a coursera course and the book is free as a pdf
“ An Introduction to Statistical Learning” even if it is with examples in R and you don’t care about R. It’s also not aimed at complete beginners, but if you have a very solid grasp on high school math and some blood on your teeth, you can totally do it. There’s also a coursera course and the book is free as a pdf
“ An Introduction to Statistical Learning” even if it is with examples in R and you don’t care about R. It’s also not aimed at complete beginners, but if you have a very solid grasp on high school math and some blood on your teeth, you can totally do it. There’s also a coursera course and the book is free as a pdf
ISLR is a great introduction to machine learning, not necessarily to statistics. It doesn't cover stuff like hypothesis testing or simulation methods.
Could probably go through it again but I’d also like more of a theoretical background.
Isn’t OLS a specific case of MLE, for example ?
Anyway, I’d like to understand machine learning and statistical inference more deeply.
I’d be happy to get the answers by reading a handful of books.