Read on Twitter
People ask me about getting started with data science all the time. So I came up with three paths for self study: Easy, Medium and Hard.
Continue reading to hear about each path and see my book recommendations.
Continue reading to hear about each path and see my book recommendations.
EASY PATH. On this path, you will get lots of practical skills but only an intuitive sense of the theory. I think the easiest way to start is to read a good data science book and get your hands dirty. I like to recommend "R for Data Science" for that.
MEDIUM PATH. This path gives you some theory with an eye toward applications. A lot of books at this difficulty level suck. The formatting and writing are often bad and the authors are clearly phoning it in. I like "Think Stats" because it's not like that.
HARD PATH. This is the hardest but the most complete path. First, you should master at least one college class worth of material in:
- Calculus
- Linear Algebra
- Probability Theory
- A statistical programming language like R
Here are a few books I like
- Calculus
- Linear Algebra
- Probability Theory
- A statistical programming language like R
Here are a few books I like
Next I would try one of these books to get a solid foundation in statistics. Both are challenging! (I like Casella & Berger personally.)
I'm sure there are many other ways of learning stats on your own so please don't feel offended if none of this reflects your experiences. I'm partly writing this as a reference for people who want to hear my personal take.
