Maarten van Smeden

@MaartenvSmeden

Statistician, assist. professor • prediction, measurement, medical statistics, epidemiology •

Utrecht
mvansmeden.net
Vrijeme pridruživanja: prosinac 2010.
502 fotografije ili videozapisa Fotografije i videozapisi

Tweetovi

Blokirali ste korisnika/cu @MaartenvSmeden

Jeste li sigurni da želite vidjeti te tweetove? Time nećete deblokirati korisnika/cu @MaartenvSmeden

  1. prije 3 sata

    The biggest difference between statistics and machine learning may be in language! So a few months ago I created this (inspired by ) but haven't made much progress since. Welcoming suggestions for improvements

    10 replies 65 proslijeđenih tweetova 228 korisnika označava da im se sviđa
    Poništi
  2. prije 4 sata

    Andrew has written some smart things about when and what to trial again. A thread worth reading

    Odgovor korisnicima @ADAlthousePhD @goldblatchikens i sljedećem broju korisnika: 3
    Sticking with our running injury example above, if you believe that there are one or more specific factors which *may* differentiate patients who benefit more from gait retraining versus load management, we can design a trial to randomized patients and then separately...
    4 proslijeđena tweeta 9 korisnika označava da im se sviđa
    Poništi
  3. proslijedio/la je Tweet
    4. velj

    People are very happy to adopt treatments without proper RCTs. But when an RCT suggests a treatment they use does not work, they suddenly become very interested in methodological rigour.

    25 replies 179 proslijeđenih tweetova 839 korisnika označava da im se sviđa
    Poništi
  4. proslijedio/la je Tweet
    prije 10 sati

    Strong knowledge on biology was translated into a prediction model. This relatively simple model is based on sufficient sample size and may impact medical care, while continuously being improved with new markers and statistical updating

    ‘A Urine-Based Genomic Assay to Triage Patients with Hematuria for Cystoscopy’ in - cc
    3 proslijeđena tweeta 17 korisnika označava da im se sviđa
    Poništi
  5. proslijedio/la je Tweet
    prije 24 sata

    Since we're sharing the good news regarding I also wanna mention that and yours truly entry ""Improving simulation studies - Towards a firm foundation for analytical decisions" has also been accepted. Gosh, SIPS2020 is gonna be LIT 🔥🔥🔥

    3 proslijeđena tweeta 46 korisnika označava da im se sviđa
    Poništi
  6. proslijedio/la je Tweet
    4. velj
    Odgovor korisniku/ci

    You haven't tried Advanced Bayesian Imputation where you replace a missing value by the value that maximizes your subjective belief

    4 proslijeđena tweeta 63 korisnika označavaju da im se sviđa
    Poništi
  7. 4. velj

    We need less peer review invitations, better peer review invitations, and peer review invitations for the right reasons -- no I don't want to make a new profile and I don't want to review your microbiome-precision-medicine-biomarker-machinelearning-AI article

    Got my old institutional e-mail address back after it has been inactive for 2 years. First e-mail? Peer review invitation...
    1 reply 5 proslijeđenih tweetova 35 korisnika označava da im se sviđa
    Poništi
  8. 4. velj

    Got my old institutional e-mail address back after it has been inactive for 2 years. First e-mail? Peer review invitation...

    47 korisnika označava da im se sviđa
    Poništi
  9. proslijedio/la je Tweet
    2. velj

    So is here a part of a thing I'm working on that I'm sharing a bit prematurely, largely to help myself find typos and just to feel like I've "done" something. Any comments are very welcome.

    42 proslijeđena tweeta 158 korisnika označava da im se sviđa
    Prikaži ovu nit
    Prikaži ovu nit
    Poništi
  10. 2. velj

    A good time to share again two good resources on p values

    12 proslijeđenih tweetova 22 korisnika označavaju da im se sviđa
    Prikaži ovu nit
    Prikaži ovu nit
    Poništi
  11. 2. velj

    The only correct probability interpretation of the p value: probability <0.05 that the given definition of the p value in a randomly selected textbook is correct

    I've been looking for a correct definition of p values in textbooks... and no wonder everyone has trouble with this. Most of the definitions are straight up wrong. Here follows a list of incorrect examples:
    Prikaži ovu nit
    4 proslijeđena tweeta 37 korisnika označava da im se sviđa
    Prikaži ovu nit
    Prikaži ovu nit
    Poništi
  12. proslijedio/la je Tweet
    1. velj

    Nice article by Michael Wallace on behalf of the in communicating clearly about measurement error concepts, consequences and potential solutions

    1 reply 15 proslijeđenih tweetova 32 korisnika označavaju da im se sviđa
    Poništi
  13. 1. velj

    Nice article by Michael Wallace on behalf of the in communicating clearly about measurement error concepts, consequences and potential solutions

    1 reply 15 proslijeđenih tweetova 32 korisnika označavaju da im se sviđa
    Poništi
  14. 1. velj

    Reality check: categorization of estimated probabilities as "low", "mediocre" and "high" based on the 25th and 75th percentiles, always a bad idea?

    7 replies 2 proslijeđena tweeta 14 korisnika označava da im se sviđa
    Poništi
  15. proslijedio/la je Tweet
    1. velj

    How to do stepwise regression in R?

    5 replies 45 proslijeđenih tweetova 148 korisnika označava da im se sviđa
    Poništi
  16. proslijedio/la je Tweet
    31. sij

    Definitely worth a read for those interesting in IPD meta-analysis. No systematic difference (on average across multiple IPD MA projects) between results from IPD and non-IPD studies, but difference may be large in any particular IPD MA project

    Free access to our article on data availability bias in meta-analysis here:
    6 proslijeđenih tweetova 13 korisnika označava da im se sviđa
    Poništi
  17. proslijedio/la je Tweet
    1. velj

    Causality & Complexity is the theme of VVSOR's Annual Meeting on March 12, 2020 . Early bird registration (new this year) has just opened. Visit the website for more info on the program and registration:

    8 proslijeđenih tweetova 10 korisnika označava da im se sviđa
    Poništi
  18. 1. velj

    Undergrad: follow this diagram to choose your stats, SPSS is your friend. Grad: SPSS is just a childhood friend, now do R. Look at these amazing complex features! Post grad receiving peer review: WHY DIDN'T YOU CONSULT A STATISTICIAN FOR YOUR COMPLEX ANALYSES?

    5 replies 26 proslijeđenih tweetova 179 korisnika označava da im se sviđa
    Poništi
  19. proslijedio/la je Tweet
    31. sij
    Odgovor korisnicima

    Ha! I'll bite. My controversial opinion would indeed be the exact opposite: we should worry more about variance/error and less about bias in epidemiology

    3 proslijeđena tweeta 10 korisnika označava da im se sviđa
    Poništi
  20. 31. sij

    I hope this is one big joke, but it probably isn't. Can't think of a more lazy way to make up data and pretend it is real. (I think you'll like this example )

    So is +45 (from a 3rd paper). He even left the formula in the .xls file on Dryad. Going through, you can see that some individual's recordings were boosted by +45 to increase within-group variability, in the trt groups for which the authors predicted high variability
    Prikaži ovu nit
    1 reply 2 proslijeđena tweeta 9 korisnika označava da im se sviđa
    Poništi

@MaartenvSmeden još nije objavio nijedan Tweet.

Čini se da učitavanje traje već neko vrijeme.

Twitter je možda preopterećen ili ima kratkotrajnih poteškoća u radu. Pokušajte ponovno ili potražite dodatne informacije u odjeljku Status Twittera.

    Možda bi vam se svidjelo i ovo:

    ·