Jonathan Bartlett

925 posts

Jonathan Bartlett banner
Jonathan Bartlett

Jonathan Bartlett

@TheStatsGeek

Biostatistician with interests in missing data, causal inference, clinical trials. Professor in Medical Statistics at @LSHTM

Bath, England Katılım Mayıs 2014
431 Takip Edilen3K Takipçiler
Jonathan Bartlett retweetledi
Triad sou.
Triad sou.@triadsou·
G-formula with multiple imputation for causal inference with incomplete data. Jonathan W Bartlett, Camila Olarte Parra, Emily Granger, Ruth H Keogh, Erik W van Zwet, Rhian M Daniel. Statistical Methods in Medical Research. journals.sagepub.com/doi/full/10.11…
Română
1
9
36
2.3K
Jonathan Bartlett retweetledi
Michael Schomaker
Michael Schomaker@mf_schomaker·
Our paper on applying missingness DAGs to longitudinal trial data has just been published in Biostatistics: academic.oup.com/biostatistics/… A challenging but rewarding experience. We give ideas on useful rules of thumbs & how to explore deviations from the assumed missingness process.
Michael Schomaker@mf_schomaker

Causal inference with missing data? Are missingness DAGs and graphical M(N)AR concepts, as proposed by @Carthica & @yudapearl, readily applicable to complex longitudinal studies? Check out our preprint led by @a_holovchak 👇 arxiv.org/abs/2402.14562

English
0
6
26
2.1K
Jonathan Bartlett retweetledi
Covariate Adjustment ASA BIOP Working Group
Covariate Adjustment ASA BIOP Working Group@CAR_SWG·
Our next journal club is scheduled on Sep 13 at 11am EST. Speaker: Kelly Van Lancker, Ghent University Zoom link: umich.zoom.us/j/7573650566 Title: Automated, efficient and model-free inference for randomized clinical trials via data-driven covariate adjustment
English
0
2
13
2.5K
Jonathan Bartlett retweetledi
Lu Mao
Lu Mao@LuMaoWisc·
Win ratio has been widely used, but so far with little attention to its estimand. I argue for articulating the estimand as a first step to analysis, and explain how👇 journals.sagepub.com/doi/10.1177/17…
English
3
8
37
5.9K
Chris O'Doherty
Chris O'Doherty@chris_odoc·
@OrionBroker9 @OctopusEnergy Wouldn’t be concerned if it were that simple!….Octopus use the app to schedule charging times at a lower rate. Guess you’re not as clever as you try to be behind your keyboard!
English
1
0
1
92
Chris O'Doherty
Chris O'Doherty@chris_odoc·
@OctopusEnergy can’t access the app or website. Any advice on how long it will be down?…Got to get the car charged!!!
English
0
0
0
79
Jonathan Bartlett
Jonathan Bartlett@TheStatsGeek·
@CatchTwentyToo My guess is that you shouldn't view them as independent events-a patch of rain is moving across an area, and the uncertainty is mostly when will it rain and not if? And to calculate the prob of a dry day you'd need to assume/know something about the dependence of hourly outcomes.
English
1
0
1
34
Dr Kristian Brock
Dr Kristian Brock@CatchTwentyToo·
So my conclusion is that the advertised probs apply not to an hour but to a whole day. I could believe, for example, that it rains on 30% of days like today, if the whole day endured conditions like those prevailing now. (I get that changes but that’s fine; the probs change)
English
1
0
0
90
Dr Kristian Brock
Dr Kristian Brock@CatchTwentyToo·
A question for statisticians, probabilists, and fans of the weather. Do you regard these probs as the probability that it will rain in each hour? Or the chances that it would rain over the course of the day if the conditions stayed as advertised?
Dr Kristian Brock tweet media
English
2
0
0
334
Jonathan Bartlett retweetledi
Christina Yap
Christina Yap@ChristinaBYap·
Exciting job opportunity for a talented statistician to join us at the Early Phase & Adaptive Trials Group @ICR_CTSU as a Trials Methodologist! If you enjoy developing and implementing efficient trial methods to impact patients’ lives, come and join us! tinyurl.com/2fpad6z2
English
0
9
19
2K

Keşfet

@SofiaSVillar1 @CAUSALab @karolinskainst @jonathanasterne @BristolUni @LSHTM_datastats @DrPerpo @RuthHKeogh