It's Tutorial Tuesday! Join Drs. Zihang Lu and Kuan Liu for a hands-on tutorial on longitudinal data analysis. Learn methods for clustering longitudinal features and causal inference with time-varying exposures, supported by real-world case studies and R code. #ENAR2025
@Zihang_Lu and I have an open CANSSI postdoctoral position. If you are interested in Bayesian statistics with applications in clinical research, feel free to reach out to us to learn more!
📢APPLY! @Zihang_Lu and I have proposed a project to work on developing new Bayesian methods for longitudinal EHR. The postdoc will also work with Dr. David Stephens from McGill. I will widely share the position next week, at the meantime see https://t.co/3eKZhFVblY for details.
My 2nd “written by witten” column is out. Thanks to @InstMathStat for giving me a platform. Also, I hope the fact that it was published on April 1 doesn’t detract from the gravitas of this piece!
https://t.co/wWpJaTFAum
I am often asked how it is that I can talk openly about my work & myself on here, without being afraid of negative feedback.
This is an important question, and the answer is: practice.
A short thread on how I’m learning to embrace critique and criticism.
Heterogeneity: all 4 of these countries have similarly advanced healthcare systems, and all had similar sized (measured) epidemics this morning (21,000-26,000 known cases).
#SARS2019p CFR
Interesting how much Italy is an outlier. Lots of testing, age structure, not overwhelm healthcare system, public health response, & intangibles seem to lead to the lower cluster of 0.5-1.5%
In this local COVID-19 talk I gave yesterday, I provided background, summarized and interpreted the main findings of our analysis of 25,000+ cases in Wuhan. I provided several take home messages and practical recommendations, especially for handling the situation in US
PLEASE RETWEET: Jacob Bien and I are hiring a post-doc!!!! It's the perfect set-up:
✅live in sunny Los Angeles
✅work on interesting statistical methodology problems
✅collaborate with Jacob and me!!
Submit your application via the link by MARCH 15!!!
https://t.co/EXHsxFx6nz
A waking-up reflection: Deep learning without deep understanding may get us into deep trouble.
(Re)Introducing HDSR, starting from the summary:
Data Science: An Artificial Ecosystem
https://t.co/ZPnZQ4YTDF
Are you going on the CS job market this year? @NicolasPapernot and I put together a (hopefully) comprehensive guide to navigating the process. Including lots of resources & examples!
https://t.co/YT8F82sdRh
From the first issue of the HDSR Radu Craiu (@rvcraiu) shares the highs and very highs of searching for and hiring in @UofTStatSci that elusive Data Scientist who has her/his feet firmly and expertly planted in two fields.
https://t.co/gWNJJJ8O30
My advice for those applying to various new investigator awards along with a couple of my applications from the NIH New Innovator (DP2) and Searle Scholar Awards. [1/6] https://t.co/pTSbohRhdr
An interesting study reported in @nature based on analysis of more
than 65 million papers, patents and software products, showing that
smaller teams tend to produce disruptive new ideas and concepts, whereas larger teams tend to develop existing ones.
https://t.co/d1c6BsHNuC
A rapid rise (from 25 to 60% since the 1960s) of scientists who spend their entire career only as supporting authors without having led a publication. The fraction of entering researchers who achieve full careers has diminished. The class of temporary scientists has escalated.