Ensemble Model

2,940 Views | 13 Replies | Last: 5 yr ago by Beat40
Fitch
User Profile
Private Message
Ignore User Stop Ignoring
How long do you want to ignore this user?
24 hours One week Permanently Cancel
11:38p, 5/18/20
AG
If anyone listened to the webcast with Dr's Peter Hotez, Rebecca Fischer and Lauren Ancel Meyers last week there was a mention the CDC has started using an ensemble approach with a dozen or so of the models that have been developed to forecasting the viral growth / contraction.

Thought it was interesting given the criticism of model forecasts through this episode...not unlike the weather.

Collated models can be found here: https://reichlab.io/covid19-forecast-hub/
Reply Quote
fightingfarmer09
User Profile Ignore User Stop Ignoring
How long do you want to ignore this user?
24 hours One week Permanently Cancel
In reply to Fitch • 7:25a, 5/19/20
So they are weathermen now?

Have you ever tracked the accuracy of a 14 day forecast?

Good for them on doing something that should've been utilized months ago. It doesn't change the human influence and garbage data that goes into the data.
Reply Quote
1 edit
Keegan99
User Profile
Private Message
Ignore User Stop Ignoring
How long do you want to ignore this user?
24 hours One week Permanently Cancel
7:50a, 5/19/20
AG
Out of curiosity, did any of these models predict what has transpired in Georgia over the past month?

https://dph.georgia.gov/covid-19-daily-status-report
Reply Quote
1
Fitch
User Profile
Private Message
Ignore User Stop Ignoring
How long do you want to ignore this user?
24 hours One week Permanently Cancel
9:11a, 5/19/20
AG
One of the better individual models in the ensemble has been from covid19-projections.com

You can see how the Georgia projection evolved day by day.
Reply Quote
5
NASAg03
User Profile Ignore User Stop Ignoring
How long do you want to ignore this user?
24 hours One week Permanently Cancel
11:02a, 5/19/20
A month later, there's no doom and gloom that EVERYONE predicted for Georgia. Hmmmmm...
Mike Shaw - Class of '03
Reply Quote
2
NASAg03
User Profile Ignore User Stop Ignoring
How long do you want to ignore this user?
24 hours One week Permanently Cancel
11:08a, 5/19/20
It's infuriating that a month ago everyone predicted Georgia's demise for opening too early, yet yesterday they had ONE covid-19 related death in the entire state.

Yet there are no headlines stating that Georgia is a success and maybe this entire thing is overblown.

edited to add a new article from the LA Times indicating that Georgia looks like a success only because they are faking their numbers, but they aren't a success and shouldn't be followed. This is beyond asinine.

https://www.latimes.com/opinion/story/2020-05-18/georgia-coronavirus-numbers-reopening-manipulated-data-brian-kemp
Mike Shaw - Class of '03
Reply Quote
6
1 edit
Fitch
User Profile
Private Message
Ignore User Stop Ignoring
How long do you want to ignore this user?
24 hours One week Permanently Cancel
11:15a, 5/19/20
AG
*An OpEd
Reply Quote
2
Keegan99
User Profile
Private Message
Ignore User Stop Ignoring
How long do you want to ignore this user?
24 hours One week Permanently Cancel
In reply to NASAg03 • 11:36a, 5/19/20
AG
NASAg03 said:
It's infuriating that a month ago everyone predicted Georgia's demise for opening too early, yet yesterday they had ONE covid-19 related death in the entire state.

Yet there are no headlines stating that Georgia is a success and maybe this entire thing is overblown.

edited to add a new article from the LA Times indicating that Georgia looks like a success only because they are faking their numbers, but they aren't a success and shouldn't be followed. This is beyond asinine.

https://www.latimes.com/opinion/story/2020-05-18/georgia-coronavirus-numbers-reopening-manipulated-data-brian-kemp


That entire OpEd is based on one bad graph. It's nothing but a brazenly partisan hack, trying to paint a picture of more death and disease for political gain.

No one seems to be questioning the graph at https://dph.georgia.gov/covid-19-daily-status-report, which is uniformly good news.
Reply Quote
3
1 edit
murphyag
User Profile
Private Message
Ignore User Stop Ignoring
How long do you want to ignore this user?
24 hours One week Permanently Cancel
12:47p, 5/19/20
AG
I have a lot of friends in Atlanta. A few that are business owners. Life is not back to normal there by a long shot. More places are open, but they aren't getting near the foot traffic they need to be in order to be profitable. Lots of people are continuing to stay home and not be out at restaurants or doing a lot of shopping.
Reply Quote
1
1 edit
Ranger222
User Profile
Private Message
Ignore User Stop Ignoring
How long do you want to ignore this user?
24 hours One week Permanently Cancel
6:00p, 5/19/20
AG
Do we need a refresher on exponential growth?
Reply Quote
1
texan12
User Profile Ignore User Stop Ignoring
How long do you want to ignore this user?
24 hours One week Permanently Cancel
In reply to Ranger222 • 6:13p, 5/19/20
Reply Quote
1
1 edit
GAC06
User Profile Ignore User Stop Ignoring
How long do you want to ignore this user?
24 hours One week Permanently Cancel
In reply to texan12 • 6:19p, 5/19/20
AG
Two more weeks
Reply Quote
3
BiochemAg97
User Profile
Private Message
Ignore User Stop Ignoring
How long do you want to ignore this user?
24 hours One week Permanently Cancel
In reply to murphyag • 6:40p, 5/19/20
AG
murphyag said:
I have a lot of friends in Atlanta. A few that are business owners. Life is not back to normal there by a long shot. More places are open, but they aren't getting near the foot traffic they need to be in order to be profitable. Lots of people are continuing to stay home and not be out at restaurants or doing a lot of shopping.
And likely what will happen across much of the country. Lots of people will continue to stay home. Lots of people will go out less than before (1-2 places a day rather than spend the day at the mall).

This is also part of the reason why the models predicting failure once the lockdowns were lifted are flawed. There is a built in assumption that once the imposed limit goes away, things will return to "normal". On top of that, it seems some models are basing on increased mobility, but few models factor in things like mask usage and maintaining 6 ft of social distancing. You can't just take the data from early March and plug that in as R once the lockdown is lifted.
Reply Quote
7
texan12
User Profile Ignore User Stop Ignoring
How long do you want to ignore this user?
24 hours One week Permanently Cancel
In reply to BiochemAg97 • 6:55p, 5/19/20
Reply Quote
1
1 edit
Beat40
User Profile Ignore User Stop Ignoring
How long do you want to ignore this user?
24 hours One week Permanently Cancel
In reply to texan12 • 7:10p, 5/19/20
texan12 said:
I doubt many were staying home, especially this past month. Have you seen an increase in Walmart, Lowe's, or Home Depot parking lots? People who were bored and went to bars before the pandemic are the same ones who are bored and go to the essential businesses.
His point is there are many more people staying home than going out as was before March, which I agree with.

I personally know several families that are not keen on gathering with other families until June 1 at minimum. They want to wait to see what's going on.

I think the first two, or three months even, a lot of people are going to be much more cautious and levels will not be at where they were before March.
Reply Quote
2
Refresh
Page 1 of 1
 
×
subscribe Verify your student status
See Subscription Benefits
Trial only available to users who have never subscribed or participated in a previous trial.