Log In

Your email is not your username

Register

If you were a member of the old Bodybuilding.com forums and would like to reuse your previous username, you can request it below. We use your email only for registration and do not store it. For more information, please see our Privacy Policy.

Confirm your email

A registration code was sent to your email. Enter it here.

Welcome

You have successfully setup your account.

Sign in

Quick Navigation Bottom Misc
Forum » Uses and Abuses of Regression Modeling
  1. Results 1 to 25 of 25
  2. Page 1 of 1

Thread: Uses and Abuses of Regression Modeling

post 1699412893 03-19-2024, 07:12 AM
-
#1
  1. J.L.C.
  2. Wat
  1. J.L.C.
  2. Wat
  3. Join Date: Apr 2005
  4. Posts: 33,220
  5. Rep Power: 37920

Uses and Abuses of Regression Modeling

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
post 1699412913 03-19-2024, 07:12 AM
-
#2
  1. messomi
  2. ▇ █ █ ▇ █ █ ▇
  1. messomi
  2. ▇ █ █ ▇ █ █ ▇
  3. Join Date: Apr 2017
  4. Posts: 20,365
  5. Rep Power: 362412
negged
post 1699414823 03-19-2024, 08:13 AM
-
#3
  1. r0bo
  2. hi friend
  1. r0bo
  2. hi friend
  3. Join Date: Mar 2007
  4. Location: Tampa, Florida, United States
  5. Posts: 9,257
  6. Rep Power: 78322
Nice try, but nobody in this cesspool cares to learn how to critically analyze a paper. They just want to bolster their ****ty, biased viewpoints
post 1699416133 03-19-2024, 08:58 AM
-
#4
  1. Dave22reborn
  2. Cold Hearted SOB
  1. Dave22reborn
  2. Cold Hearted SOB
  3. Join Date: Jan 2005
  4. Location: Ill.
  5. Posts: 120,479
  6. Rep Power: 316531
Is that why you still make excuses for people who mask up?
post 1699416333 03-19-2024, 09:02 AM
-
#5
  1. J.L.C.
  2. Wat
  1. J.L.C.
  2. Wat
  3. Join Date: Apr 2005
  4. Posts: 33,220
  5. Rep Power: 37920
Originally Posted By Dave22reborn⏩
Is that why you still make excuses for people who mask up?
I don't think wearing a mask is wrong, so no excuses are needed.

But, this paper might help with developing some literacy
post 1699418103 03-19-2024, 09:38 AM
-
#6
  1. J.L.C.
  2. Wat
  1. J.L.C.
  2. Wat
  3. Join Date: Apr 2005
  4. Posts: 33,220
  5. Rep Power: 37920
Originally Posted By r0bo⏩
Nice try, but nobody in this cesspool cares to learn how to critically analyze a paper. They just want to bolster their ****ty, biased viewpoints
You can lead a horse to water, but you can't make it drink
post 1699418163 03-19-2024, 09:40 AM
-
#7
  1. frankdtank20
  2. Registered User
  1. frankdtank20
  2. Registered User
  3. Join Date: Nov 2020
  4. Posts: 27,464
  5. Rep Power: 171374
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.
Yeah Buddyyy! Light weight! Light weight baby!!!!
post 1699418343 03-19-2024, 09:45 AM
-
#8
  1. J.L.C.
  2. Wat
  1. J.L.C.
  2. Wat
  3. Join Date: Apr 2005
  4. Posts: 33,220
  5. Rep Power: 37920
Originally Posted By frankdtank20⏩
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.
Maybe. I haven't seen the model you're referencing. Can you share it?

More like misinterpreting conditioning and coefficient estimates to draw incorrect conclusions through table 2 fallacies
post 1699419153 03-19-2024, 10:06 AM
-
#9
  1. Dave22reborn
  2. Cold Hearted SOB
  1. Dave22reborn
  2. Cold Hearted SOB
  3. Join Date: Jan 2005
  4. Location: Ill.
  5. Posts: 120,479
  6. Rep Power: 316531
Originally Posted By J.L.C.⏩
I don't think wearing a mask is wrong, so no excuses are needed.

But, this paper might help with developing some literacy
You don't think that people who still mask up are ignorant? Do you still mask up?
post 1699419183 03-19-2024, 10:06 AM
-
#10
  1. Dave22reborn
  2. Cold Hearted SOB
  1. Dave22reborn
  2. Cold Hearted SOB
  3. Join Date: Jan 2005
  4. Location: Ill.
  5. Posts: 120,479
  6. Rep Power: 316531
Originally Posted By frankdtank20⏩
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.
Trust the experts!!!
post 1699419193 03-19-2024, 10:07 AM
-
#11
  1. J.L.C.
  2. Wat
  1. J.L.C.
  2. Wat
  3. Join Date: Apr 2005
  4. Posts: 33,220
  5. Rep Power: 37920
Originally Posted By Dave22reborn⏩
You don't think that people who still mask up are ignorant? Do you still mask up?
I said I didn't think it's wrong. If it's not wrong, there's nothing to "excuse".

Regardless, what does this or Frank's non-sequitor have to do with regression modeling?
post 1699419903 03-19-2024, 10:26 AM
-
#12
  1. frankdtank20
  2. Registered User
  1. frankdtank20
  2. Registered User
  3. Join Date: Nov 2020
  4. Posts: 27,464
  5. Rep Power: 171374
Originally Posted By Dave22reborn⏩
Trust the experts!!!
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.
Yeah Buddyyy! Light weight! Light weight baby!!!!
post 1699420113 03-19-2024, 10:31 AM
-
#13
  1. J.L.C.
  2. Wat
  1. J.L.C.
  2. Wat
  3. Join Date: Apr 2005
  4. Posts: 33,220
  5. Rep Power: 37920
Originally Posted By frankdtank20⏩
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.
Can you share the regression model and maybe link it to the abuses/mistakes referenced in the OP?
post 1699420493 03-19-2024, 10:38 AM
-
#14
  1. DuracellBunny
  2. Has new batteries!
  1. DuracellBunny
  2. Has new batteries!
  3. Join Date: Nov 2005
  4. Posts: 15,594
  5. Rep Power: 201320
Originally Posted By frankdtank20⏩
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.
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......
post 1699420683 03-19-2024, 10:42 AM
-
#15
  1. ezmac31
  1. ezmac31
  3. Join Date: Jun 2009
  4. Location: Michigan, United States
  5. Age: 43
  6. Posts: 20,160
  7. Subscribers: 1
  8. Rep Power: 247931
Originally Posted By r0bo⏩
Nice try, but nobody in this cesspool cares to learn how to critically analyze a paper. They just want to bolster their ****ty, biased viewpoints
You sound vaccinated and quadruple boosted
post 1699420753 03-19-2024, 10:43 AM
-
#16
  1. J.L.C.
  2. Wat
  1. J.L.C.
  2. Wat
  3. Join Date: Apr 2005
  4. Posts: 33,220
  5. Rep Power: 37920
Originally Posted By DuracellBunny⏩
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.
Did they use regression?
post 1699420843 03-19-2024, 10:45 AM
-
#17
  1. Dave22reborn
  2. Cold Hearted SOB
  1. Dave22reborn
  2. Cold Hearted SOB
  3. Join Date: Jan 2005
  4. Location: Ill.
  5. Posts: 120,479
  6. Rep Power: 316531
Originally Posted By J.L.C.⏩
I said I didn't think it's wrong. If it's not wrong, there's nothing to "excuse".

Regardless, what does this or Frank's non-sequitor have to do with regression modeling?
Why isn't it wrong for someone to still mask up? What's the reasoning, especially since they're negative side effects.
post 1699420913 03-19-2024, 10:46 AM
-
#18
  1. J.L.C.
  2. Wat
  1. J.L.C.
  2. Wat
  3. Join Date: Apr 2005
  4. Posts: 33,220
  5. Rep Power: 37920
Originally Posted By Dave22reborn⏩
Why isn't it wrong for someone to still mask up? What's the reasoning, especially since they're negative side effects.
Can you share the regression model you're referencing?
post 1699421203 03-19-2024, 10:52 AM
-
#19
  1. r0bo
  2. hi friend
  1. r0bo
  2. hi friend
  3. Join Date: Mar 2007
  4. Location: Tampa, Florida, United States
  5. Posts: 9,257
  6. Rep Power: 78322
Originally Posted By ezmac31⏩
You sound vaccinated and quadruple boosted
You sound upset you can't comprehend the topic at hand
Maybe try the circle jerk ctard threads with your fellow dimwits
post 1699424703 03-19-2024, 12:07 PM
-
#20
  1. tyqb4
  2. Registered User
  1. tyqb4
  2. Registered User
  3. Join Date: Nov 2009
  4. Age: 34
  5. Posts: 4,461
  6. Rep Power: 12989
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
post 1699425093 03-19-2024, 12:16 PM
-
#21
  1. J.L.C.
  2. Wat
  1. J.L.C.
  2. Wat
  3. Join Date: Apr 2005
  4. Posts: 33,220
  5. Rep Power: 37920
Originally Posted By tyqb4⏩
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.
This is why empirical data are best analyzed using scientifically informed generative models
"all models are wrong, but some models are useful" is a great quote
Agreed!
post 1699426493 03-19-2024, 12:52 PM
-
#22
  1. Dave22reborn
  2. Cold Hearted SOB
  1. Dave22reborn
  2. Cold Hearted SOB
  3. Join Date: Jan 2005
  4. Location: Ill.
  5. Posts: 120,479
  6. Rep Power: 316531
Originally Posted By r0bo⏩
You sound upset you can't comprehend the topic at hand
Maybe try the circle jerk ctard threads with your fellow dimwits
You individuals who are still masking up, and getting every jab possible are truly enlightened.....
post 1699426833 03-19-2024, 01:04 PM
-
#23
  1. r0bo
  2. hi friend
  1. r0bo
  2. hi friend
  3. Join Date: Mar 2007
  4. Location: Tampa, Florida, United States
  5. Posts: 9,257
  6. Rep Power: 78322
Originally Posted By Dave22reborn⏩
You individuals who are still masking up, and getting every jab possible are truly enlightened.....
I'd say you should join that other retard in the circlejerk threads, but someone knocked over a mannequin at hot topic, mall cop. Back to work!
post 1699437783 03-19-2024, 06:46 PM
-
#24
  1. Dave22reborn
  2. Cold Hearted SOB
  1. Dave22reborn
  2. Cold Hearted SOB
  3. Join Date: Jan 2005
  4. Location: Ill.
  5. Posts: 120,479
  6. Rep Power: 316531
Originally Posted By r0bo⏩
I'd say you should join that other retard in the circlejerk threads, but someone knocked over a mannequin at hot topic, mall cop. Back to work!
Hmmmmmmm, all of those booster shots must lead to an increase in estrogen, because you sure are acting like a bitch.
post 1699437843 03-19-2024, 06:48 PM
-
#25
  1. ezmac31
  1. ezmac31
  3. Join Date: Jun 2009
  4. Location: Michigan, United States
  5. Age: 43
  6. Posts: 20,160
  7. Subscribers: 1
  8. Rep Power: 247931
Originally Posted By r0bo⏩
You sound upset you can't comprehend the topic at hand
Maybe try the circle jerk ctard threads with your fellow dimwits
Definitely boosted. And upset
Quick Navigation Top Misc
Bookmarks
Digg.com
Digg
del.icio.us
del.icio.us
Stumbleupon.com
StumbleUpon
Google.com
Google
Facebook.com
Facebook
Posting Permissions
  1. You may not post new threads
  2. You may not post replies
  3. You may not post attachments
  4. You may not edit your posts