#statstwitter search results
🔥New method: Free Discontinuity Regression (FDR)! (w/ Florian Gunsilius) Combines ✨multidimensional✨ changepoint estimation & nonparametric regression in one shot. Find jump locations & sizes in any dimension, w/ formal convergence guarantees🧵👇 #EconTwitter #StatsTwitter
Hey #statstwitter! This is old homeopathy trial. Sample size is 27+26. I just can't understand how unadj null/inconclusive effects become strongly significant with adjusting. Adjustment was done with variables showing stat sig imbalance at baseline 🤦. Still too good to be true?
#epitwitter #statstwitter My father-in-law is a scientist. He is insanely gifted. We were looking at a research paper together years ago and I asked him what it would cost to write it today. I will never forget his answer… "We can’t, we don’t know how to do it."
Delighted to have picked up Boen & Zahn 1982 but note the previous reader circled this section on "anger" 😟 #statstwitter
✨📰preprint - From prediction to prescription: Machine learning and Causal Inference hal.science/hal-04774700 Machine learning for individual treatment effect Didactic and precise, bridging statistical, computational, and epidemiological thinking. #statstwitter #epitwitter 1/5
Alright #statstwitter, new example of confounding variable to use in your Stat101 classes just dropped!
Hey #statstwitter I need an idiot check here: The shaded areas are bootstrapped confidence intervals. This plot makes no sense, right? Why are there bands to the left of the data? Why do the bands only go down and not up at -20 and -19? Is there anything I'm missing here?
Authors -> "The external validation AUC was 0.727 (95% CI: 0.5763–0.8779 [...]), and the consistency between the actual observation and predicted probability was good." Me ->🙄 + 😭 #stoptheBS #statstwitter #honestinterpretationplease
![GSCollins's tweet image. Authors -> "The external validation AUC was 0.727 (95% CI: 0.5763–0.8779 [...]), and the consistency between the actual observation and predicted probability was good."
Me ->🙄 + 😭
#stoptheBS #statstwitter #honestinterpretationplease](https://pbs.twimg.com/media/F1ORx1VWYAANbV-.png)
Twitter methodology peeps Immortal time bias type qu Observational study -Enrolment at day zero (propensity matching) -Intervention at D14 in one group -Follow up censored until D14 for both groups Does this introduce bias? #statstwitter #epitwitter
TONIGHT! #statstwitter and #econtwitter go toe-to-toe in the boxing match of the century. @MatthewBJane and I will be duking it out on a livestream at 6pm PST. @ChelseaParlett will announce, @Kyle_Ireton will referee. youtube.com/live/c2fIDLPfC…
It 'should' be obvious but you can't use a prediction model when the predictors aren't known at the moment of prediction. Here @pauladhiman and I comment on one such prediction model... --> tinyurl.com/5c56t6ef #statstwitter #epitwitter
Hey all you #statstwitter #statistics data wonks, I've made an online sheet that allows you to dynamically compare a linear trend to observed values. Just "save a copy" of this sheet and play with it yourself! confidence intervals, yay! docs.google.com/spreadsheets/d…
Just a reminder of some common statistical faux pas and how to avoid them in our 2022 @bmj_latest xmas paper (#notjustforxmas) Longer discussion of each issue can be found here --> tinyurl.com/4ktf7sdd #statstwitter
A teachers guide to the modelling of epidemics #MathTwitter #MathsTwitter #StatsTwitter #EpiTwitter #EduTwitter github.com/stochanswers/p…
#OpenAccess @BMC_series #Trials #statstwitter looking for the details of how we’re analysing @BOOSTICStudy? #BOOSTIC: Statistical elements and design, lead by @mdymock25 @drdavidwjg @j_h_mcmahon @MonashUni @AlfredHealth rdcu.be/exXi8
Na análise de sobrevivência, tratar riscos competitivos como censura pode gerar vieses. Neste post, mostro como considerar esses eventos com CIF, teste de Gray e modelo de Fine & Gray no R. tiagoms.com/posts/riscocom… #rstats #statstwitter #epitwitter #stats
9/10 💡 Found this thread useful? • ❤️ Like & 🔁 Retweet to share the knowledge • 💬 Comment your war stories with nonnormal errors! Let’s build better models together 🚀 #StatsTwitter #FinanceTwitter
Hypothesis testing is vital in science, business, and daily choices. It stops us from jumping to conclusions and helps us make informed decisions. #StatsTwitter #DataAnalytics 8/8
NBA player height isn't normal #statstwitter stochanswers.com/studies/height…
We often repeat similar data transformations manually — it’s error-prone and hard to maintain. Here’s a simple, adaptable example: create a new column from others using just their names. #rstats #datascience #statstwitter
In the 1998 crash GBM was seen not to be sufficient. My model uses affine returns to get a better fit. This results in a combination of GBM and another motion - a mixture distribution of the log-normal plus another. stochanswers.com/studies/financ… #MathTwitter #MathsTwitter #StatsTwitter…
We don’t change the mountain—we renovate bad houses so models can fit well. 🔄 Log ➗ Square Root 🧰 Box-Cox ➡️ Data transformation isn’t cheating. It’s clarity. Read more: rb.gy/xc3lii #Chisquares #StatsTwitter #Regression #DataTransformation
Hey #statstwitter people, what do we think about interpreting coefficients in cloglog binomial regressions? Since everyone hates odds ratios I wonder if interpreting hazard ratios is easier.
#statstwitter #epitwitter Comparing Forecasting Models for Potato Production: Evaluating T-ARMA, ARIMA-ARCH, Weibull and Score-Driven Approaches in Major Global Producers India Pakistan researchgate.net/publication/39…
FDR isn't just for #StatsTwitter #EconTwitter! Useful for #MachineLearning, #DataScientists, #climate, urban studies, #finance, gene expression, #marketing & more. If your data has jumps, FDR can help! 🚀 arxiv.org/abs/2309.14630 @EmoryEconomics @umichECON
🔥New method: Free Discontinuity Regression (FDR)! (w/ Florian Gunsilius) Combines ✨multidimensional✨ changepoint estimation & nonparametric regression in one shot. Find jump locations & sizes in any dimension, w/ formal convergence guarantees🧵👇 #EconTwitter #StatsTwitter
One nice stat by Tamil commentator : 📍Whenever Tilak Varma scored 50+ score , Mi lost their matches. #MIvDC #statstwitter #MI
#statstwitter #epidemiology #mathtwitter #mathstwitter #edutwitter Maths in the playground stochanswers.com/education/tag.…
📖🎉My new book, Impact Evaluation in Firms and Organizations With Applications in R and Python, will be published on August 5th by @mitpress and is now available for pre-order: mitpress.mit.edu/9780262552929/… #econtwitter #statstwitter
Step 2 - reviving old hashtags e.g. #statstwitter #epitwitter and tagging folk who I loved following/engaging with to see if they are still here: @jamesheathers @deevybee @sTeamTraen @MaartenvSmeden @stephensenn @GidMK @K_Sheldrick @lakens @d_spiegel @david_colquhoun
New open-access paper: Inverse Probability of Treatment Weighting Using the Propensity Score With Competing Risks in Survival Analysis. #stats #statstwitter onlinelibrary.wiley.com/doi/10.1002/si…
Anyone know how to tell if students are intentionally calling Tukey tests Turkey tests? Should we be correcting them? #newPI #StatsTwitter
🔥New method: Free Discontinuity Regression (FDR)! (w/ Florian Gunsilius) Combines ✨multidimensional✨ changepoint estimation & nonparametric regression in one shot. Find jump locations & sizes in any dimension, w/ formal convergence guarantees🧵👇 #EconTwitter #StatsTwitter
✨📰preprint - From prediction to prescription: Machine learning and Causal Inference hal.science/hal-04774700 Machine learning for individual treatment effect Didactic and precise, bridging statistical, computational, and epidemiological thinking. #statstwitter #epitwitter 1/5
Hey #statstwitter! This is old homeopathy trial. Sample size is 27+26. I just can't understand how unadj null/inconclusive effects become strongly significant with adjusting. Adjustment was done with variables showing stat sig imbalance at baseline 🤦. Still too good to be true?
NEW PAPER: ✨How do public health researchers interpret regression coefficients?✨ doi.org/10.1186/s12889… Not a survey or quiz, this is qualitative work exploring diverse epistemological and interpretative practice in biostatistics. #statstwitter #epitwitter
Halloween week is here 🎃 ! Let the demonstration of survivor bias begin! #statstwitter #statistics #teaching
Hey #StatsTwitter. Looking at a table from a trial and the standard error across different treatment groups is identical at multiple time points. Is this a common statistical occurrence or a potential error?
Hey #statstwitter, how do I do what I'm doing* with nicer code? *I'm trying to pivot_longer() multiple indexed variables where each variable gets a column.
💣Utilizar ChatGPT para analizar tus datos es fácil, pero comprender si el análisis es realmente fiable es otro cantar. Sin conocimientos estadísticos adecuados, no sabrás evitar análisis incorrectos y resultados mal interpretados.💥 Imagen: @statsepi #stats #statstwitter #AI
Delighted to have picked up Boen & Zahn 1982 but note the previous reader circled this section on "anger" 😟 #statstwitter
This children’s book features a group of women statisticians & data scientists stephaniehicks.com/superwomen/ #womeninscience #statstwitter #datascience
#epitwitter #statstwitter My father-in-law is a scientist. He is insanely gifted. We were looking at a research paper together years ago and I asked him what it would cost to write it today. I will never forget his answer… "We can’t, we don’t know how to do it."
Something went wrong.
Something went wrong.
United States Trends
- 1. Marshawn Kneeland 40.5K posts
- 2. Nancy Pelosi 57K posts
- 3. Craig Stammen 1,531 posts
- 4. Gordon Findlay 1,306 posts
- 5. #MichaelMovie 60.4K posts
- 6. Pujols N/A
- 7. Ozempic 4,707 posts
- 8. #NO1ShinesLikeHongjoong 35.9K posts
- 9. Abraham Accords 2,998 posts
- 10. #영원한_넘버원캡틴쭝_생일 35.2K posts
- 11. Eli Lilly 3,338 posts
- 12. GLP-1 3,879 posts
- 13. Baxcalibur 5,689 posts
- 14. Preller N/A
- 15. Kyrou N/A
- 16. Gremlins 3 4,612 posts
- 17. ESPN Bet 3,087 posts
- 18. Chimecho 8,374 posts
- 19. Kazakhstan 4,658 posts
- 20. Joe Dante N/A