Perhaps of interest to R&P scientists and statisticians
On the uses and abuses of regression models: a call for reform of statistical practice and teaching
"By way of our own historical examples and a brief review of mainstream clinical research journals, we describe how this perspective leads to a range of problems in the application of regression methods, including misguided "adjustment" for covariates, misinterpretation of regression coefficients and the widespread fitting of regression models without a clear purpose"
https://arxiv.org/abs/2309.06668
|
-
03-19-2024, 06:12 AM #1
Uses and Abuses of Regression Modeling
-
03-19-2024, 06:12 AM #2
-
03-19-2024, 07:13 AM #3
-
03-19-2024, 07:58 AM #4
-
-
03-19-2024, 08:02 AM #5
-
03-19-2024, 08:38 AM #6
-
03-19-2024, 08:40 AM #7
Hence tards telling the western world 65% of your daily calories should be carbs and eating more than 60-70 grams of protein per day as a man is excessive. Don't forget to eat snacks in between every meal, so your insulin is high all day. Cuz like das healthy!!!! Canola oil, an industrial lubricant, is healthy!!! Chronic inflammation, just take a pill, brah.
riiiiiiiiiiiight.Light weight! Light weight baby!!!!
-
03-19-2024, 08:45 AM #8
-
-
03-19-2024, 09:06 AM #9
-
03-19-2024, 09:06 AM #10
-
03-19-2024, 09:07 AM #11
-
03-19-2024, 09:26 AM #12
The crappy thing is that high carb diet advice came originally from a huge study (I forget the name) that only included data that agreed with the hypothesis, while half of the data was discarded. Excluded data that disagreed with the hypothesis, zero adjustment for other behavioral factors. Fraudulent AF. And it screwed the world way more than any fast spreading respiratory virus or clot shot can.
Light weight! Light weight baby!!!!
-
-
03-19-2024, 09:31 AM #13
-
03-19-2024, 09:38 AM #14
It was a harvard study paid for by the sugar producers of america, or something like that - a sugar industry lobbying group. Back then (late 60s/early 70s) there was no requirement to state who had paid for the study.
Because the study was done by 3 (I think it was 3) harvard researchers, it was accepted at face value and then cited, without anybody else doing a similar study.
So the study paid for by the sugar people concluded fat = bad and sugar = good, which then dictated policy for generations.Screw nature; my body will do what I DAMN WELL tell it to do!
The only dangerous thing about an exercise is the person doing it.
They had the technology to rebuild me. They made me better, stronger, faster......
-
03-19-2024, 09:42 AM #15
-
03-19-2024, 09:43 AM #16
-
-
03-19-2024, 09:45 AM #17
-
03-19-2024, 09:46 AM #18
-
03-19-2024, 09:52 AM #19
-
03-19-2024, 11:07 AM #20
pretty sure the way regression is taught in college is if your descriptive statistical tests are within a certain range for your model then the model is good, no considerations for anything else.
health/human body related problems seem way more complex than typical problems in other industries though. looking at a product or a process with limited variables and most things being definable and measurable is like playing a completely different game.
"all models are wrong, but some models are useful" is a great quote
-
-
03-19-2024, 11:16 AM #21
-
03-19-2024, 11:52 AM #22
-
03-19-2024, 12:04 PM #23
-
03-19-2024, 05:46 PM #24
-
-
03-19-2024, 05:48 PM #25
Bookmarks