Kristen Hunter (@stats_hunter) header
Kristen Hunter (@stats_hunter) profile picture

Lecturer in statistics & data science @UNSW | PhD from @HarvardStats | #rstats enthusiast

Sydney, New South Wales
Joined July 2019
Photo by Kristen Hunter (@stats_hunter)Photo by Kristen Hunter (@stats_hunter)Photo by Kristen Hunter (@stats_hunter)Photo by Kristen Hunter (@stats_hunter)Photo by Kristen Hunter (@stats_hunter)Photo by Kristen Hunter (@stats_hunter)Photo by Kristen Hunter (@stats_hunter)Photo by Kristen Hunter (@stats_hunter)Photo by Kristen Hunter (@stats_hunter)Photo by Kristen Hunter (@stats_hunter)
Kristen Hunter retweeted
not gonna lie it was a little disturbing to learn that a large amount of the post-apocalyptic jargon in Mad Max / Furiosa is actually just regular Australian slang
72
866
132
30,012
1,806,942
Celebrate #IWD2024 with us! We introduce you to 25 rising stars in #DataScience: 5 hours/5 women each hour 📌Online 📆Tues, 12 March 🕛12pm-5pm (MEL, SYD, CBR) 🕦11:30am-4:30 pm (ADL) 🕚11am-4pm (BNE) 🕘9am-2pm (PER) 🕘9pm, 11 Mar: USA EDT events.humanitix.com/austral… #womeninSTEM
1
14
8
32
9,239
Kristen Hunter retweeted
The day I got into data visualization, I promised myself that someday, someone would pay me to crochet a chart. That day is today. washingtonpost.com/home/2023…
12
79
16
485
44,767
Kristen Hunter retweeted
Regular reminder of the best mathematical resource in machine learning, The Matrix Cookbook. Don't know how anyone ever does any math without it. math.uwaterloo.ca/~hwolkowi/…
10
261
15
1,579
218,297
Kristen Hunter retweeted
What do you do if you want to control for baseline Xs but some have missing values? The folk wisdom is to dummy them out, something I’ve been teaching students to do for years. Now I have an actual citation for this approach!
2
16
1
87
23,799
Kristen Hunter retweeted
Can blocking hurt? It turns out the answer is "it depends," but in this case it is depending on even how we think of blocking and data, which is cool. See my esteemed collaborator Nicole Pashley's blog post on it: cares-blog.gse.harvard.edu/p…
1
7
388
Kristen Hunter retweeted
A good power analysis iterates over possible effect sizes rather than pulling one out of a hat. Or, how much power do we have to detect small, medium or large effects? Make a bunch of plots and vary a lot of things to understand different plausible counter factuals.
Hot take: Almost all power analyses are simply math hiding estimated guesses about an effect size. When it is a new experiment, there is no good way to make that estimate. So it is just math wrapped around some BS.
3
1
21
3,300
Sample size calculations become a lot easier when you realize "Hey fuck it, I can just simulate this"
10
11
5
225
Kristen Hunter retweeted
*NEW* #EGAPMethodsGuide 10 Things to Know About Multisite or Block-Randomized Trials egap.org/resource/10-things-… Thanks to author @stats_hunter for your work on this guide!
2
2
Take a look at a fun blog by Stats PhD Alum Dr. Kristen Hunter and co-author Lily An on matching and difference-in-difference methods! @LMiratrix @stats_hunter
My CARES lab (Causality, Applications, and Research in Education and Statistics) is (very slowly) starting a blog where we put various things we are thinking about regarding, well, causality, applications, & research in education and statistics! Check out cares-blog.gse.harvard.edu/
It was our pleasure to have Kristen @stats_hunter. Wonderful presentation. Thank you for your time.
4
8
Kristen Hunter retweeted
My CARES lab (Causality, Applications, and Research in Education and Statistics) is (very slowly) starting a blog where we put various things we are thinking about regarding, well, causality, applications, & research in education and statistics! Check out cares-blog.gse.harvard.edu/
2
28
2
182
The #IESfunded Power Under Multiplicity Project (PUMP) introduces new software tools for estimating power. This project is a collaboration between researchers at @MDRC_News & Harvard and is funded by @IESResearch.
1
7
8
Kristen Hunter retweeted
At @RLadiesMelb @melvidoni not only talks about the @rOpenSci #reproducibleResearch #reviewProcess and how amazing are the #ropenSci community are in reviewing packages, but also introduces the amazing #unconf happening in Australia!
1
7
13