More catchup! Feels a little cheaty to use convolve2d, but I think this is a bit prettier than manually padding the matrix.
#AdventOfCode

Day 3 - Catching up! #AdventOfCode

Day 2 of #AdventofCode

Little behind, but Day 1 of #AdventOfCode

Peep some new software done with the @causalab.bsky.social! (Now available on CRAN), SEQTaRget can be used to estimate observational analogs of intention-to-treat and per-protocol effects and can adjust for potential selection bias induced by LTFU.
Leave a star on git to track development!
#Rstats

"The cancellation of federal funding to HSPH is not only a tragedy for our colleagues and their staff and students, but it will also prove to be a tragedy for the health of our fellow citizens," read a statement signed by more than 20 retired Harvard Chan School professors.

Now available on CRAN: circles 0.1.0 ! Currently in developmental infancy, it now provides an adaptation of MATLAB’s bubblebath by Adam Danz to the R ecosystem: check it out now!

github.com/ryan-odea/ci...
#rstats #aRt #generativeArt

Harvard’s School of Public Health is hosting a hackathon to archive public health data! Take part in preservation of crucial public health data in-person (Harvard ID holders) or over zoom!

#publicHealth #Harvard #epidemiology #biostatistics

harvard.az1.qualtrics.com/jfe/form/SV_...

Showing off a new #rstats #python package in collaboration with Sean McGrath! Introducing LEARNER: Latent Space-Based Transfer Learning. We leverage similarity in row and column spaces to improve estimation in the target population.

Check it out in R as “learner” or python as “learner-py”!