Tim Morris
@timpmorris.bsky.social
Biostatistician working on methodology at Novartis. Simulation studies, non-inferiority, missing data, estimands, covariate adjustment…
He/him
https://tpmorris.substack.com/
He/him
https://tpmorris.substack.com/
You just know people will interpret 0.89 hours as 1h29m. So I’d go defensive and use HH:MM
February 3, 2026 at 4:52 PM
You just know people will interpret 0.89 hours as 1h29m. So I’d go defensive and use HH:MM
😆 as a treat. Is there a limit on the number of treats we can reward ourselves with per paper?
February 3, 2026 at 11:50 AM
😆 as a treat. Is there a limit on the number of treats we can reward ourselves with per paper?
My 11 year old has jumped quickly to the “humour” stage of grief
January 25, 2026 at 6:33 AM
Glad you spotted that small show of contempt (also comic sans)
January 16, 2026 at 1:30 PM
Glad you spotted that small show of contempt (also comic sans)
He did, just people didn’t seem to get the memo!
January 16, 2026 at 1:04 PM
He did, just people didn’t seem to get the memo!
All I really want is for people to stop with the personal-twists. Darren’s has topped the alternative-evidence-pyramids pyramid for like a decade now.
January 16, 2026 at 1:00 PM
All I really want is for people to stop with the personal-twists. Darren’s has topped the alternative-evidence-pyramids pyramid for like a decade now.
Ha are you also thinking about reconstructing the data from Kaplan–Meier figures? I don’t imagine getting the numbers out would be different to any other plot, but presume you still need a package to do the Guyot method (like ipdfc in Stata) or whatever.
January 14, 2026 at 11:23 AM
Ha are you also thinking about reconstructing the data from Kaplan–Meier figures? I don’t imagine getting the numbers out would be different to any other plot, but presume you still need a package to do the Guyot method (like ipdfc in Stata) or whatever.
I hadn’t thought about it… do you mean specifically for potential outcome simulation? Do say more.
January 9, 2026 at 4:01 PM
I hadn’t thought about it… do you mean specifically for potential outcome simulation? Do say more.
That post was written to get you interested and to read this preprint with Alex Ocampo, Enrico Giudice and Zachary McCaw on Gaussian quadrature – your best option if feasible (not always).
arxiv.org/abs/2601.05128
arxiv.org/abs/2601.05128
Revealing the Truth: Calculating True Values in Causal Inference Simulation Studies via Gaussian Quadrature
Simulation studies are used to understand the properties of statistical methods. A key luxury in many simulation studies is knowledge of the true value (i.e. the estimand) being targeted. With this or...
arxiv.org
January 9, 2026 at 1:44 PM
That post was written to get you interested and to read this preprint with Alex Ocampo, Enrico Giudice and Zachary McCaw on Gaussian quadrature – your best option if feasible (not always).
arxiv.org/abs/2601.05128
arxiv.org/abs/2601.05128
You might be interested in this paper by @andershuitfeldt.net, Robin?
ORs as a measure turn out to be disappointing.
arxiv.org/abs/2106.06316
ORs as a measure turn out to be disappointing.
arxiv.org/abs/2106.06316
Shall we count the living or the dead?
In the 1958 paper "Shall we count the living or the dead?", Mindel C. Sheps proposed a principled solution to the familiar problem of asymmetry of the relative risk. We provide causal models to clarif...
arxiv.org
January 9, 2026 at 8:16 AM
You might be interested in this paper by @andershuitfeldt.net, Robin?
ORs as a measure turn out to be disappointing.
arxiv.org/abs/2106.06316
ORs as a measure turn out to be disappointing.
arxiv.org/abs/2106.06316