@kathoffman.bsky.social biostatistician + PhD student @UWBiostat. learning + writing about better ways to conduct research. #rstats, #dataviz, causal inference
Seattle, WA
Joined January 2018
- Tweets476
- Following650
- Followers2.2K
- Likes1.6K
Pinned Tweet
A tutorial on Longitudinal Modified Treatment Policies-- a flexible method for defining, identifying, and estimating causal parameters of interest-- is now in @EpidemiologyLWW!
🔗journals.lww.com/epidem/abst…
cc🌟coauthors: @dasalazarb @nickWillyamz @kara_rudolph @ildiazm
updated tutorial on Longitudinal Modified Treatment Policies is now on arxiv!
🔗 arxiv.org/abs/2304.09460v3
for those at ACIC, i'll be hanging out by this poster today from 5-6:30pm and would love to chat about LMTPs, methodology tutorials, etc.
Kat Hoffman retweeted
Apropos of nothing, here is a problem set question from one of my undergrad econometrics courses
Kat Hoffman retweeted
Starts October 2! PhD applicants applying to @UWBiostat and @UWStat may use a pre-application review service designed for applicants from historically marginalized groups. Information: biostat.washington.edu/apply…
Kat Hoffman retweeted
What should you call your academic event?
Another great flowchart from PhD Comics.
ALT A Flowchart about naming your academic event from Jorge Cham of PhD Comics.
Kat Hoffman retweeted
Many of my students struggle with the distinction between Ordinary Least Squares (OLS) and Maximum Likelihood Estimation (MLE).
To clarify, we use an interactive #Python dashboard that fits parameters to a Gaussian distribution based on a given dataset.
This visual approach helps them see the difference: OLS minimizes the mismatch between the model and the data, while MLE maximizes the likelihood of the data given the model parameters. With this tool, the concepts become much clearer!
I share it on #GitHub @ github.com/GeostatsGuy/DataS… ∀. #DataScience #MachineLearning
Kat Hoffman retweeted
As application season rolls around again, here's your reminder that materials from my successful applications are available on my website (NSF-GRFP, Google PhD Fellowship, Stanford data science postdoc, and CS faculty job search): anjalief.github.io/statement…
Kat Hoffman retweeted
Great Idea for Poster presentation...!!!
Via Caroline O'Donnell/O'Flaherty Lab (Linkedin)
#scicomm #scienceoutreach #sciencetalk #scienceposter
amazing resource for PhD applicants by @drlucylai below!
tag-teaming off her with my own post on my Biostatistics PhD application experience (applied Fall '22) ⤵️
khstats.com/blog/phd-apps/
✨it's PhD application season again!✨ i am re-sharing my popular guide on applying for PhD programs in STEM, that has now been viewed 60k+ times over the past 6 years😍 please RT and share with anyone that might find it useful! lucylai.com/blog/gradapps
Kat Hoffman retweeted
I can't remember the last time I renamed variables manually in #rstats. I always use my data dictionary for renaming! That way, if I ever need to update names in the future, I'm only updating once in my data dictionary, not in both my data dictionary and script. 🙌
Kat Hoffman retweeted
We’re seeking a new chair to lead a dedicated team of faculty, staff, and students who are passionate about developing and using rigorous quantitative methods to improve the well-being of communities in the United States and around the world. bit.ly/3MmZHwE
Thank you to @AmstatNews for the support, and so many mentors for guidance/encouragement.
Extra thanks to @ildiazm for helping me start methods research pre-PhD.
Looking forward to year 2 at @UWBiostat!!
Congrats to @uwbiostat PhD student Kat Hoffman who has received the Gertrude M. Cox Scholarship. The award recognizes Hoffman’s methodological research developing machine-learning based methods drawing on causal inference on longitudinal interventions. bit.ly/4fHQOv8
thank you @BhramarBioStat for years of encouragement, going all the way back to when I was an undergrad-- giving me an invitation to attend summer BDSI talks as soon as you learned I lived locally 💛💙
Michigan will miss you!
You’re unable to view this Post because this account owner limits who can view their Posts. Learn more
Kat Hoffman retweeted
theres definitely some overlap in the R packages {lmtp} and {ltmle}! here's some diffs:
- lmtp can be used for MTP estimands (intervening on natural value of treatment)
- lmtp encodes an additional estimator (SDR) for static/dynamic/MTPs
- lmtp does not encode MSMs
updated tutorial on Longitudinal Modified Treatment Policies is now on arxiv!
🔗 arxiv.org/abs/2304.09460v3
for those at ACIC, i'll be hanging out by this poster today from 5-6:30pm and would love to chat about LMTPs, methodology tutorials, etc.
new tutorial paper! 🤩 check it out if you're looking to add a general causal inference method to your toolbox. 🧰
LMTP + #rstats pkg {lmtp} is especially useful for time-varying data and/or continuous exposures.
🔗 : arxiv.org/pdf/2304.09460.pdf
pre-print highlights ⬇️ 1/n
Kat Hoffman retweeted
Wondering why you can't find the 13th floor? Madison Hardesty and @jauerbach_stats find out how much New Yorkers are prepared to pay not to live on the “unlucky” level
ow.ly/9PG250Ryh7L
#freetoread #statistics
Kat Hoffman retweeted
When asked to draw a scientist, school-age kids in the U.S. are increasingly sketching women, according to a study from 2018. #ScienceMagArchives scim.ag/4Nt
I'll be a panelist on this virtual discussion about applying to (bio)statistics phds, taking place in a few minutes! it will be recorded if you can't make the live session ⤵️
Join us for a virtual conversation and Q&A with current Statistics Ph.D. students from the Departments of Statistics at UW, UC Berkeley, and UMich Ann Arbor on Wed., Nov. 1, 2023 at 5:00 PM PT / 8:00 PM ET. See details here: bit.ly/3tCqYVZ
