Author Topic: COVID-19 Data Model Updated Daily  (Read 63101 times)

Beasho

  • Cortez Bank Status
  • *****
  • Posts: 3224
    • View Profile
Re: COVID-19 Data Model Updated Daily
« Reply #210 on: April 11, 2020, 09:06:03 AM »
This article is the most complete I have seen on the collection of Obesity related complications to Covid-19:

https://www.businessinsider.com/americas-obesity-epidemic-may-make-the-coronavirus-pandemic-worse-2020-4

1) "a "sophisticated internist" in New York City told him at least 90% of the people under age 50 he's seen ventilated due to COVID-19 were obese. "

2) Fatima Fakhoury, a weight-inclusive registered dietitian in New York, told Business Insider a colleague in Cornell's surgical intensive-care unit, told her "every one" of her 60 ventilated patients in a single day, whether 28 or 85, was obese.

3) Looking at 383 COVID-19 patients in China, researchers found that people who were overweight had an 86% higher likelihood or developing severe pneumonia than their normal weight counterparts. Those who were obese were 2.4 (read 240%) times more likely to develop it.

4) Higher-weight men were particularly at risk, with those who overweight almost twice as likely to develop the condition and those who were obese 5.7 (read 570%) times more likely.  "Those are really stunning odds ratios," Dr. Gregory Poland, an infectious disease expert at the Mayo Clinic, told Business Insider.

5) "Prior to COVID-19, we already knew that obesity is a risk factor for infection in general and more severe complications, and this is particularly true for respiratory infections," Poland said.   During the 2009 H1N1 epidemic in 2009, adults with body mass indexes over 30 were 3.1 (read 310%) times more likely to die from the infection than people of lower weights, while adults with body mass indexes over 40 were 7.6 (read 760%) times more likely to die.

6) "At the immunological system level, the human body is extremely poorly adapted to chronic caloric overload," Poland said. "It's as bad for our immune system as malnutrition is."  Particularly among people who carry their body fat around the middle, obesity can activate inflammation and weaken the body's anti-inflammatory response, making them more susceptible to illness, research suggests.

« Last Edit: April 11, 2020, 09:21:29 AM by Beasho »

Beasho

  • Cortez Bank Status
  • *****
  • Posts: 3224
    • View Profile
Re: COVID-19 Data Model Updated Daily
« Reply #211 on: April 11, 2020, 09:10:55 AM »
I have been screaming about CoMorbidity since I saw the statistics 3 weeks ago out of China then confirmed by Bloomberg in Italy and the data from Imperial College.  All 100% aligned on the risk to PRIOR HEALTH CONDITIONS.

Then on Sunday 60 Minutes was reporting and the doctor said "We see a high correlation with Obesity." BMI > than 40 is listed as Morbid Obesity.  And there it was the word again.

Obesity.  Might the Corona virus just target Obesity. 

Obesity as the Leading indicator of death NOT as the Last of a short list of CoMorbidities.

Obesity, Heart Disease, Cancer, Respiratory Disease . . .


PS: The President knows this.  He was at a press conference 3 weeks ago citing how terrible the disease is.  He said something like "I have a friend with the virus.  One day he is fine 3 days later he is in a coma on a ventilator.  It's vicious . . He was a little heavy . . but its terrrible."  He knows.

---------
Reuters on Obesity:
“This virus is terrible, it can hit young people, in particular obese young people. Those who are overweight really need to be careful,” Delfraissy told France info radio.
https://www.reuters.com/article/us-health-coronavirus-france-confinement/obesity-is-major-covid-19-risk-factor-says-french-chief-epidemiologist-idUSKBN21Q0S7

World Obesity at Risk with BMI >> 25
 “Given the extremely high rates of obesity around the globe we expect that a high percentage of the population who will contract coronavirus will also have a BMI over 25.”
https://www.worldobesity.org/news/statement-coronavirus-covid-19-obesity

Doctors on Obesity (in Connecticut):
Dr. John Morton who heads up Bariatric Surgery with Yale Medicine explains, “There’s something called hypoventilation syndrome and that is when you have extra tissue around your chest.  It’s harder to take deep breaths.”
Obesity also increases the risk for health issues such as diabetes and high blood pressure.
https://www.wtnh.com/news/health/coronavirus/what-you-need-to-know-about-obesity-and-covid-19/

Scott Pelley: Have you found a common denominator for why some patients crash so precipitously?

Dr. Mangala Narasimhan: We do think that there's some trends towards obesity, that patients who are obese seem to do-- worse and men definitely more than women.

Forward to 9:15 in the Report here:
https://www.cbsnews.com/news/new-york-city-coronavirus-epicenter-united-states-peak-60-minutes-2020-03-29/

Admin

  • Administrator
  • Cortez Bank Status
  • *****
  • Posts: 6443
    • View Profile
    • StandUpZone
    • Email
Re: COVID-19 Data Model Updated Daily
« Reply #212 on: April 11, 2020, 09:35:19 AM »
The model presented in this tool has a 95% uncertainty interval and is represented by the shaded area(s) on each chart.[/b][/size]

This really pisses me off.  I work to make sure my reports are near 100% perfect.  There are a few small math issues with running totals and averages based upon how the data comes in but I would be willing to say my data is 98%+ accurate with regards to the source vs. 95% wrong.  Holy Moly.

Hi Beasho,

They are talking about the unreliable nature of the underlying data itself, not internal errors in their reports or math issues.  Testing data, start dates, etc.  It is a real mess.  Data is never perfect but this is really remarkably bad. 

To make that even more notable, that is without any changes to any of the parameters.  That is a different issue altogether.  Change (for instance) the value given to distancing and that would be an entirely different model snapshot.   



« Last Edit: April 11, 2020, 09:42:14 AM by Admin »

LaPerouseBay

  • Teahupoo Status
  • ******
  • Posts: 1969
  • downwind dilettante
    • View Profile
    • Email
Re: COVID-19 Data Model Updated Daily
« Reply #213 on: April 11, 2020, 10:56:23 AM »
^ Thanks Admin.


I’m hoping for a 30% infection rate in CA right now as I type. It might not be there but a guy can hope.

Prevalence in Germany at ~15%.  Dr. Attia's spreadsheet was an eye opener for me - valuable info for a layperson like myself as the US moves forward with testing.   

https://youtu.be/FxA2-vl9kdw
« Last Edit: April 11, 2020, 10:58:21 AM by LaPerouseBay »
Support your local shaper

LaPerouseBay

  • Teahupoo Status
  • ******
  • Posts: 1969
  • downwind dilettante
    • View Profile
    • Email
Re: COVID-19 Data Model Updated Daily
« Reply #214 on: April 11, 2020, 06:07:55 PM »
^ Examples of the numbers he mentioned early in the video.

Also, 15% and 30% at the numbers in his POC kit.   

 
« Last Edit: April 11, 2020, 06:11:09 PM by LaPerouseBay »
Support your local shaper

Beasho

  • Cortez Bank Status
  • *****
  • Posts: 3224
    • View Profile
Re: COVID-19 Data Model Updated Daily
« Reply #215 on: April 12, 2020, 11:58:21 AM »

There may be a link to people who suffer from hypertension.

Found this for NY while searching for deaths by age group, but not sure if it exists on a national level.  Breaks out the comorbidities as well as age groups

https://covid19tracker.health.ny.gov/views/NYS-COVID19-Tracker/NYSDOHCOVID-19Tracker-Fatalities

More data on CoMorbidity and it is practically hidden.  I had to stitch together 3 Screen shots to put this view together.  I am not kidding you can NOT see this view anywhere.  You have to scroll through 3 rows at a time.

81% of Deaths in NY listed as CoMorbid.

Yet again numbers Closer to 100% than even 50% EVERYWHERE YOU TURN.  Obesity is NOT listed as a CoMorbidity.

Whoops Data updated.  Now NY is showing 88% CoMorbid related Deaths. 

I woke up this morning and read some articles that relating Covid-19 to the Spanish Flu.  NO COMPARISON!

Anyone suggesting this is like the Spanish flu is trying to push an agenda or terrorize.  Other than Social Isolation and wearing Masks the two diseases are like comparing basketball to sewing.

1)   The Spanish Flu killed Healthy people age 20 to 30.
2)   Covid-19 is Killing Obese People with CoMorbidites Age 75+
3)   Young people are Not Dying who are not Obese or CoMorbid.  They are safer by a factor of 10X to 50X than from common influenza
4)   The Spanish Flu killed between 50 and 100 Million people
5)   Covid-19 will not likely kill 1% as many people as the Spanish flu.

The following graphic shows the age of fatalities comparing the 1918 Pandemic vs. Covid-19 as measured by the State of NY midnight April 11th.  Representing nearly 10% of the World’s deaths the data shows a Peak Age group = 70-79 years.  The Spanish flu attacked young healthy people age 20 to 30.  In NY State the most recent data showed 88% of the Case Fatalities involved CoMorbidity.  NY does NOT list Obesity as a Comorbid condition which could push the number closer to 100%.



all~wet

  • Rincon Status
  • ***
  • Posts: 233
    • View Profile
Re: COVID-19 Data Model Updated Daily
« Reply #216 on: April 12, 2020, 01:10:50 PM »
@Beasho- Appreciate the raw data. 

Whether it is over dramatizing and sensationalizing to prop up an agenda / scare people into complicity or minimizing what is really going on to provide false comfort or promote self interest- most of the info we are fed is at best being manipulated or just plain inaccurate.

Quite refreshing and useful to treat people as adults- enable everyone to see unvarnished truth/unfiltered numbers and let them be the judge. Most of the time, you shine light on- take a good long look in the face of most of our fears- they lose a good bit of thier steam.  Anyway- thanks for the tool to do that with updated numbers.

lucabrasi

  • Teahupoo Status
  • ******
  • Posts: 1813
    • View Profile
Re: COVID-19 Data Model Updated Daily
« Reply #217 on: April 12, 2020, 10:24:24 PM »
Thanks for taking the time to post all this Beesho.

I saw this thread and damn if I was going to look at it...…...and didn't till today.
Looked through the first 3 or so pages and the last 3 or so......going to go through it all sometime soon.
I had literally (yes, literally) terrified myself for a week or so....would not watch any news and barely peaked at online news for a bit as well.
Pretty much stayed away from anything covid.
It's real shit...but...….

Been wondering about a few of the "things" in here....and just wondering.
some random thoughts/observations.

Saw some mass grave in NY. It's used for no next of kin/unkown like people.
I saw a stat that said usually 25 a week. Up to 24 a day here recently. How long, how often, etc....I don't know but happening "now". Think it was linked to the bbc. I read a few sentences.

Drones in Savannah Georgia doing the social distancing legwork for law enforcement. Something like out of some futuristic movie.

I see the "California had it earlier" gaining traction. People around here have really wondered the same thing for a few weeks. Vegas as well I hear.

There is an article in The New Yorker on Sun Valley. Not a bad skim.

Blaine County was coming out with high numbers when testing kicked in. It had flattened out (I think) last I noticed but was still off from rest of state/country=per capita cases.
Roughly 23,00 people and 452 cases. (Idaho was one of the last states to report any cases)
Perfect breeding dish...gondolos, trams, lines all over, crowded bars and restraunts…..
I'm sure there are lots of differences in the proportion of tests given and all sorts of other stuff....

Anyways, lot's of people in Teton County think it was here in January as well. Vegas as well I hear.
I had been curious about the resort areas and the cases in those areas so I looked some up by their counties.

Jackson Hole = same population (53)
Steamboat = same population (36)
Vail = twice as big (452)
Park City = twice as big (68)
Aspen = 3/4 smaller (49)
Crested Butte = 3/4 smaller (101)
Mammoth = smallest of them all at 14,000 (21)
Vegas = 2,000,000 (2,813)

Have no clue what any of that means or if it means anything, especially with all the differences in the amount of tests done, this, that, and everything else about all this...….but I had been wondering about resort areas and how they were faring and what they were seeing,

Be curious to see these antibody tests done on the people who thought they had it earlier in the year.
I know NYC and it's world travelers, it's density, etc. etc. but I think California and Vegas would have had as much exposure to something like this just as early. Vegas has got to be just as bad of a breeding dish as anywhere else in the US.

I knew the numbers would be skewed cuz of the testing and well, all sorts of stuff.
I knew they would go off the charts at some point because of the testing.....
and I still scared the shit out of myself.

Ya'll ever hear of Doctor Steve or his show Weird Medicine?
He's on all that social media stuff.
A sane voice out there.

Thanks again for taking the time to post this Beesho.
I knew it would be.....what it is...a place to get good info....I just couldn't look till today.
Be interesting to see how it has evolved....


 


LaPerouseBay

  • Teahupoo Status
  • ******
  • Posts: 1969
  • downwind dilettante
    • View Profile
    • Email
Re: COVID-19 Data Model Updated Daily
« Reply #218 on: April 13, 2020, 06:46:59 PM »
 Maui is getting a bit weird.

Here's an Oahu doctor that has organised a team of volunteers to come to the smaller islands and test.  Feds and State seem to be mired in the mud.  His comments are not what we want to hear about our hospital management.  It seems they have been lying to staff about "No PPE available"  This doctor called 'em out.  Petitions are starting.  I'm hoping Tulsi 'heads on pikes' Gabbard takes a swing, with a flamethrower.   

https://mauinow.com/2020/04/12/video-dr-scott-miscovich-shares-front-line-view-of-covid-19-testing-in-maui-county/

Two more videos up on testing.

https://youtu.be/QmcnWqsVfdg

https://youtu.be/kgzFAdYwYLM

 

   
Support your local shaper

WaveWashed Hatteras

  • Waikiki Status
  • *
  • Posts: 19
  • Perspective, use it or lose it
    • View Profile
    • My Wife's Gallery- Ocean / Beach Art
Re: COVID-19 Data Model Updated Daily
« Reply #219 on: April 13, 2020, 07:55:16 PM »
The model presented in this tool has a 95% uncertainty interval and is represented by the shaded area(s) on each chart.[/b][/size]

This really pisses me off.  I work to make sure my reports are near 100% perfect.  There are a few small math issues with running totals and averages based upon how the data comes in but I would be willing to say my data is 98%+ accurate with regards to the source vs. 95% wrong.  Holy Moly.

Hi Beasho,

They are talking about the unreliable nature of the underlying data itself, not internal errors in their reports or math issues.  Testing data, start dates, etc.  It is a real mess.  Data is never perfect but this is really remarkably bad. 

To make that even more notable, that is without any changes to any of the parameters.  That is a different issue altogether.  Change (for instance) the value given to distancing and that would be an entirely different model snapshot.

I stumbled onto this thread and started reading it in reverse when I saw the exchange above.   Is this tongue in cheek sarcasm between Admin and Beasho or do those two really not understand how accurate a 95% uncertainty level represents?  If I missed the intent I apologize in advance.   

If for example a manufacturer said that it's new 14' Raceboard was guaranteed to have a dry weight of 14.3kg +/- 0.2kg with a 95% uncertainty then out of a production run of 100 boards the following would be true.   68 boards would weigh between 14.2 and 14.4 kg, 95 boards would weigh between 14.1 and 14.5kg and 99.7 would weigh between 14.1 and 14.5 kg.   Except for the constants of nature one would never achieve 100% accuracy.    These confidence levels or uncertainty levels only apply when the sample set is Gaussian.  In this case does the glasser use the corrrect ratio of resin to cloth?    The statement 95% uncertainty does not indicate that only 5 out of 100 boards would be in the 14.1 to 14.5 kg range.

This thread seems to be arguing, and I fully agree with argument, that Bayesian probability needs to be employed to understand this virus.   Take the simple example below;

"A patient goes to see a doctor. The doctor performs a test with 99 percent reliability--that is, 99 percent of people who are sick test positive and 99 percent of the healthy people test negative. The doctor knows that only 1 percent of the people in the country are sick. Now the question is: if the patient tests positive, what are the chances the patient is sick?"     The intuitive answer is 99 percent,

but the correct answer is 50 percent...."    Only one out of 100 patients is actually sick, the test gets every 100th result wrong.   Combine them and the risk of the getting a false result is just a coin flip.   

This virus is not killing indiscriminately, other factors are having a huge influence on the outcome of the few are carriers.   Explaining this concept to the masses is more difficult than ensuring the few who read this post understand the general terms used in typical probability statements.  We should be happy that even though these comorbidity numbers are not in the mainstream they are being published and available for anyone who searches.   We are lucky to live in a country that is so open to providing scientific results to its people via the web, even if the talking heads in the media don't take the time to report them.
Sunova Speeed 9'5"
Sunova Kanga 10'
Exocet 12'6" Downwind
C4 XXL iSup (pinhole leak at seam)
Black Project Pure Paddle
Flying Fish Carbon Paddle

LaPerouseBay

  • Teahupoo Status
  • ******
  • Posts: 1969
  • downwind dilettante
    • View Profile
    • Email
Re: COVID-19 Data Model Updated Daily
« Reply #220 on: April 13, 2020, 10:18:47 PM »
Tip of the hat to WaveWashedHatteras, Beasho and Admin for your input.  All of you have a better grasp on the math than I do.  I'm just tagging along, hoping to learn a thing or two.

The youtube link I posted on the 11th seems to have gone away.  Here it is again.

https://youtu.be/d2z73eTFvTM

I've also inserted 1% prevalence, 99% specificity and 99% sensitivity into Dr. Attia's spreadsheet, simply as a thought exercise to go along with WWH's post. 
Support your local shaper

Admin

  • Administrator
  • Cortez Bank Status
  • *****
  • Posts: 6443
    • View Profile
    • StandUpZone
    • Email
Re: COVID-19 Data Model Updated Daily
« Reply #221 on: April 14, 2020, 01:39:00 AM »
I stumbled onto this thread and started reading it in reverse when I saw the exchange above.   Is this tongue in cheek sarcasm between Admin and Beasho or do those two really not understand how accurate a 95% uncertainty level represents?  If I missed the intent I apologize in advance.   

Hi Wavewashed,

They make it easy for us.  The IMHE posted the definition below and gave us a chart (and have represented uncertainty in the huge shaded area).

The shaded areas that they mention in their definition currently shows a potential USA death range between 180,000 people and 30,000 people.  150,000 deaths is a huge range of uncertainty.  The quote that I had posted (pasted again below) from the IMHE sums it up nicely in terms that anyone can easily understand.  They include that definition and explanation on every one of their charts.

Also worth noting that when we discussed this originally the IMHE projected deaths in the USA had been lowered to 60,000.  The projected US deaths have now been raised back up to 69,000 by August 4th.  That is again assuming full social distancing through May 2020.


"Shaded area indicates uncertainty"

"Uncertainty is the range of values that is likely to include the correct projected estimate for a given data category. Larger uncertainty intervals can result from limited data availability, small studies, and conflicting data, while smaller uncertainty intervals can result from extensive data availability, large studies, and data that are consistent across sources.  The model presented in this tool has a 95% uncertainty interval and is represented by the shaded area(s) on each chart."


COVID-19 projections assuming full social distancing through May 2020
Last updated April 13, 2020 (Pacific Time).
https://covid19.healthdata.org/united-states-of-america

« Last Edit: April 14, 2020, 06:56:39 AM by Admin »

Beasho

  • Cortez Bank Status
  • *****
  • Posts: 3224
    • View Profile
Re: COVID-19 Data Model Updated Daily
« Reply #222 on: April 14, 2020, 07:19:48 AM »
Tip of the hat to WaveWashedHatteras, Beasho and Admin for your input.  All of you have a better grasp on the math than I do.  I'm just tagging along, hoping to learn a thing or two. . . .
The youtube link I posted on the 11th seems to have gone away.  Here it is again.

https://youtu.be/d2z73eTFvTM


I went to this guys website (Peter Attia) and downloaded the spreadsheet.  I have to admit it is making my hair hurt.   

I feel like my wife and kids when they say

"Dad I have no idea what your are talking about . .  no one understands your graphs and you're just saying blah, blah, blah . . $&^%^&&^"

I respond "But it is so obvious." 

Clearly not obvious.

On this link is a great podcast by Amash Adalja.  He is super level headed speaks what I would consider the unvarnished truth and concludes that the fatality rate is like 0.3% to at most 1.0%. 

Me speaking - We are lucky because this was a dry run at what could be much worse.

https://peterattiamd.com/covid-19/

WaveWashed Hatteras

  • Waikiki Status
  • *
  • Posts: 19
  • Perspective, use it or lose it
    • View Profile
    • My Wife's Gallery- Ocean / Beach Art
Re: COVID-19 Data Model Updated Daily
« Reply #223 on: April 14, 2020, 10:15:53 AM »
I stumbled onto this thread and started reading it in reverse when I saw the exchange above.   Is this tongue in cheek sarcasm between Admin and Beasho or do those two really not understand how accurate a 95% uncertainty level represents?  If I missed the intent I apologize in advance.   


The shaded areas that they mention in their definition currently shows a potential USA death range between 180,000 people and 30,000 people.  150,000 deaths is a huge range of uncertainty.  The quote that I had posted (pasted again below) from the IMHE sums it up nicely in terms that anyone can easily understand.  They include that definition and explanation on every one of their charts.


When I read your post it seems like you understand Uncertainty but then when you comment that 150,000 deaths is a huge range I wonder if there is complete understanding of how this data is presented.   Numbers are fun, lets play some statistics game with them.
IMHE says 30,000 to 180,000 with 95% uncertainty, this is not a Gaussian distribution since the number of deaths can't be negative, but the projection is in the second third of the projection, or fairly weighted.   The best part of uncertainty is that there is an equal probability of any outcome in the model.,   e.g unlike Vegas where the house has the advantage on the roulette wheel with the 0 and 00 spaces, the IMHE model gives an equal chance to every value between 30,000 and 180,000.   We can be optimists or pessimists when viewing these projects but the math says either outcome or any in-between outcome is equally likely.

Let's use a real number set from https://covid19.healthdata.org/united-states-of-america  projected for May 1

51,405 deaths with 95% uncertainty that this number will be between  28,282 and 112,584   
 
lets do a simple conversion to 99.7%

51,405 deaths with 99.7% uncertainty that this number will be between 16,721 and 153,047

the last conversion would be for 68%

51,405 deaths with 68% uncertainty that this number will be between 40,000 and 60,000  ( I rounded here for simplicity)

Again equal likelihood based on stated uncertainty of any value within the range occurring.  So since today's actual reported death toll is 22,000 the model is flawed as the 99.7% estimate is low.  One would expect that actual value would not be lower than the estimated value. My guess is that as we progress in time that IMHE will be modifying their estimates as new data is collected.     Since I'm a scientist in the spectroscopy field and not infectious diseases I need to find some perspective to determine if these uncertainties are in fact "quite huge" or actually normal.   Looking at the CDC website and their projected death rate for Influenza in the USA for Winter/Spring 2019/2020 they listed an estimate of 24,000 to 62,000 deaths with a 95% uncertainty were expected.   They note that these numbers were generated from a pool of data that represents 8.5% of the US population (~27 million people)   So the CDC has a sample pool that is many magnitudes larger than the world wide Covid19 pool and has been studying this disease extensively for the past 60 years and the best guess of deaths has a  range of 38,000.   For a novel virus like SARS V2 with near zero historical references, a death estimate with a range of 150,000 or only 3x that of  one of the most highly studies infectious diseases in the world does not seem unreasonably large.

Now take into account China under reported deaths, rumors are putting the actual Chinese death toll between 20,000 and 1,000,000 and our estimates don't seem out of line.  Then consider that if 62,000 Americans had died of the flu in the period from November to today, not a single bit of the economy would have been shuttered, I'd be competing in the OBX-Wind Long Distance and Slalom races this week, and life would be going on as normal, instead of social distancing on the farm for the past two + weeks.   Current flu deaths for USA are in the 23,000 range and current CoVID19 in the 22,000 range, for a total of 45,000 which is approximately the Mid Point of the CDC Flu estimate ( 43,000 +/- 20,000 K=2) Every death is tragic, but in the big picture we're having a relatively mild Influenza season and must be doing something right to be keeping our CoVID19 numbers so low. 

This thread's deep analysis is what I wish I'd be viewing on the mainstream media's news reports, not something buried in a Standup zone forum.   

Sunova Speeed 9'5"
Sunova Kanga 10'
Exocet 12'6" Downwind
C4 XXL iSup (pinhole leak at seam)
Black Project Pure Paddle
Flying Fish Carbon Paddle

PonoBill

  • Cortez Bank Status
  • *****
  • Posts: 25864
    • View Profile
Re: COVID-19 Data Model Updated Daily
« Reply #224 on: April 14, 2020, 11:53:37 AM »
Thanks wavewashed, I wrote a similar comment but didn't post it because I chickened out on defending my generally weak understanding of statistical math. I have to retake the Kahn Academy math classes every few years to recharge my understanding since it's so fragile (and my memory is so bad) that anything I don't use disappears.

One of the problems is the replacement of the term "Confidence" with "Uncertainty". The meaning of the terms is the same. Saying 95% uncertainty does NOT mean that the error band is 95 percent. It literally means that 95 times out of a hundred the actual number will be within the error band. The problem here is that the term "uncertainty" (or confidence if you like that better) is being used without the error band.

If your data sucks you make the error band wide enough to have 95% uncertainty (confidence) and determining that is just math. 95% uncertainty is generally the gold standard for data analysis. A higher confidence level does not necessarily mean better data, it often means that the error band is wider for a given set. No one says 100 percent confidence unless they are full of shit or the error band encompasses all possible outcomes, including "we get hit by an asteroid".
« Last Edit: April 14, 2020, 11:58:01 AM by PonoBill »
Foote 10'4X34", SIC 17.5 V1 hollow and an EPS one in Hood River. Foote 9'0" x 31", L41 8'8", 18' Speedboard, etc. etc.

 


SimplePortal 2.3.7 © 2008-2024, SimplePortal