Just a warning to anyone using G*power for power analysis of multiway ANOVA - results can vary wildly depending on setting of options, and the defaults may not be what you want. Lakens pointed this out years ago, but I suspect many people, like me, are not aware. see https://pubpeer.com/publications/5329C2DD6804C773DB3BD25DC8380C#6 for example of 3-4 fold difference in N depending on options - plus references to useful resources #statistics
I was reminded again today how much my lack of knowledge of #statistics impedes my ability to do, or even understand, certain kinds of science. It's literally my biggest academic regret that I never took a course in stats as part of my Maths degree.
So my advice to any kids out there wanting to work in the sciences (including social sciences): LEARN STATISTICS!
Polster Pradeep Gupta Of Axis My India Exit Poll Fame Breaks Down On Live TV After being teased by host that his prediction of 400 seats was very wrong
“[Josh] Gibson is MLB’s new all-time career leader in batting average (.372, moving ahead of Ty Cobb), slugging percentage (.718, moving ahead of Ruth), OPS (1.177, ahead of Ruth), and holds the all-time single season records in each of those categories”
(He still can’t top Barry Bonds’ .609 OBP in 2004 though!)
First was a great talk by Jordan Ellenberg on the importance of uncertainty and contradiction in math and why mathematics is actually part of the humanities at the Santa Fe Institute https://www.youtube.com/watch?v=L1pXQNaS9Oo (2/11) #math#statistics
This is for fans of #NoMansSky and #statistics I decided to turn my gameplay into a potential inferential statistics lesson. Nanite farming specifically. This is what I discovered feeding Delicious Vegetable Stew to Cronus in the Nexus. Yes. I repeatedly fed him stew 325 times, and wrote down 168 of those attempts in sequence. Here is the result of that experiment. I should point out that twice in all my attempts I received 329 nanites but that occurred outside these 168 recorded trials.
(1/2) Shiny Apps for demystifying statistical models and methods 🚀
This is a cool website that explains different statistical concepts with the use of interactive Shiny Apps. Ben Prytherch made this website from the Department of Statistics at Colorado State University.
For model calibration (esp via logistic regression), does anyone know of a statistical investigation of the properties of the resulting calibrated predictions?
IOW, if we use predictions from one model as inputs to another model, do we know the probability distribution of the final predictions?
I don't think many of my stats folks are here, but FYI - I am registered for this year's Joint Statistical Meetings! Hope to see a bunch of friends there