Effective Altruism Forum
EA Forum

Hide table of contents
Comment Permalink
Answer by geoffrey1
0
0

Hi Khai, this depends on what you want to do in the future. The short answer is no. Both statistics and maths are broad fields with solid generalizability and respectability. They also tend to vary a bit in difficulty, rigor and focus across schools.

Math is prob better for keeping the option of various fields of academia open. Stats is prob better for industry. But it’ll depend on the classes you take too.

The most generalizable classes will be:

  • Calculus Sequence
  • Linear Algebra
  • intro to probability and statistics

These are used in a very wide range of fields. But after that it branches out pretty quickly and you want to focus on domain knowledge or technical classes specific to a field

Economics has its own approach to stats called econometrics which deviates quite a bit culturally and technically in its focus. Andrew Gelman has some blog posts you can search on that

Stuff I don’t know which other stats people are more likely to know:

  • Markov chains
  • Monte Carlo simulations
  • really any simulation technique
  • Bayesian stats
  • information theory
  • textual analysis or ML stuff

…and a lot more. And I can / will learn a few of these in the future for work or interest. But they’re not immediately useful

Reply
See in context
by Transient Altruist
1 min read3 answers 0

2

Career choiceEducationStatistics
Frontpage

More specifically I am talking about undergraduate study.

The impression I am getting from the internet is that maths is considered harder and better - with the same benefits as stats, plus more options. You can move from maths to stats but not as easily from stats to other areas of maths.

However when it comes to having an impact, is it right to say that the maths stuff that you wouldn't also do in a stats course is not directly useful? In this case it would be better to just focus on stats, which is all applicable. Or would a stats course actually close options for having impact compared to maths?

(If it makes any difference I will be studying economics too)

(Obviously how good I would be and how much I would enjoy each course are important, but I'll consider those factors separately)

Thanks!

2

0
0

Reactions

0
0
New Answer
New Comment

3 Answers sorted by
Top

Jaime Sevilla

6
1
0

My sense is that pure math courses mostly make it easier to do pure math later, and otherwise not make much of a difference. In terms of credentials, I would bet they are mostly equivalent - you can steer your career later through a master or taking jobs in your area of interest.

Statistics seems more robustly useful and transferable no matter what you do later. Statistics degrees also tend to cover the very useful parts of math as well, such as calculus and linear algebra.

Radical Empath Ismam

4
1
0

Absolutely not. If anything, it is the other way around.

Statistics is much more applied. I did a statistics degree and became an actuary. It has been a very rewarding and impactful career. I am often quite shocked at the poor data/ statistics skills demonstrated by maths graduates, who (presumably because they are specialised in subjects like topology) are far behind their graduate peers with backgrounds in Psychology or Economics.

geoffrey

1
0
0

Hi Khai, this depends on what you want to do in the future. The short answer is no. Both statistics and maths are broad fields with solid generalizability and respectability. They also tend to vary a bit in difficulty, rigor and focus across schools.

Math is prob better for keeping the option of various fields of academia open. Stats is prob better for industry. But it’ll depend on the classes you take too.

The most generalizable classes will be:

  • Calculus Sequence
  • Linear Algebra
  • intro to probability and statistics

These are used in a very wide range of fields. But after that it branches out pretty quickly and you want to focus on domain knowledge or technical classes specific to a field

Economics has its own approach to stats called econometrics which deviates quite a bit culturally and technically in its focus. Andrew Gelman has some blog posts you can search on that

Stuff I don’t know which other stats people are more likely to know:

  • Markov chains
  • Monte Carlo simulations
  • really any simulation technique
  • Bayesian stats
  • information theory
  • textual analysis or ML stuff

…and a lot more. And I can / will learn a few of these in the future for work or interest. But they’re not immediately useful

More from Transient Altruist
View more
Curated and popular this week
Relevant opportunities
View more