Because rolling 3 day averages will show if there are true spikes because of the lag in reporting.
The Coronavirus, not the beer
- Thread starter Runnin
- Start date
Because rolling 3 day averages will show if there are true spikes because of the lag in reporting.
nsacpi
Expects Yuge Games
are the weekend numbers different than the weekdays? If so a 3 day average is distorted by that.
are the recent data incomplete and revised later? If so a 3 day average will also be distorted by that.
nsacpi
Expects Yuge Games
28,000 uncounted deaths
https://www.nytimes.com/interactive/2020/04/21/world/coronavirus-missing-deaths.html
At least 28,000 more people have died during the coronavirus pandemic over the last month than the official Covid-19 death counts report, a review of mortality data in 11 countries shows — providing a clearer, if still incomplete, picture of the toll of the crisis.
In the last month, far more people died in these countries than in previous years, The New York Times found. The totals include deaths from Covid-19 as well as those from other causes, likely including people who could not be treated as hospitals became overwhelmed.
These numbers undermine the notion that many people who have died from the virus may soon have died anyway. In Paris, more than twice the usual number of people have died each day, far more than the peak of a bad flu season. In New York City, the number is now four times the normal amount.
The differences are particularly stark in countries that have been slow to acknowledge the scope of the problem. Istanbul, for example, recorded about 2,100 more deaths than expected from March 9 through April 12 — roughly double the number of coronavirus deaths the government reported for the entire country in that period.
https://www.nytimes.com/interactive/2020/04/21/world/coronavirus-missing-deaths.html
At least 28,000 more people have died during the coronavirus pandemic over the last month than the official Covid-19 death counts report, a review of mortality data in 11 countries shows — providing a clearer, if still incomplete, picture of the toll of the crisis.
In the last month, far more people died in these countries than in previous years, The New York Times found. The totals include deaths from Covid-19 as well as those from other causes, likely including people who could not be treated as hospitals became overwhelmed.
These numbers undermine the notion that many people who have died from the virus may soon have died anyway. In Paris, more than twice the usual number of people have died each day, far more than the peak of a bad flu season. In New York City, the number is now four times the normal amount.
The differences are particularly stark in countries that have been slow to acknowledge the scope of the problem. Istanbul, for example, recorded about 2,100 more deaths than expected from March 9 through April 12 — roughly double the number of coronavirus deaths the government reported for the entire country in that period.
Last edited:
It is still going to show the peaks are lower as time progresses. There will be 3 days periods that have no weekends.
nsacpi
Expects Yuge Games
If you compare a 3 day average with a prior 3 day average, you data will be distorted by the weekend effect every single day.
You can compare 3 day averages of similar periods [tues-thurs].
This is not hard to assess the trends that infections are dropping.
nsacpi
Expects Yuge Games
yes that would not be distorted by the weekend effect
I still think a seven day average over the prior seven days is better...just because the daily data have a lot of noise in them...so the bigger the moving average the better
A rolling would show the volatility. Ultimately comparable periods will show declines the past 2 weeks
Weekend data doesn't start reflecting until Tuesday/Wednesday, so even those dates are distorted.
No it doesn't. Last week is even more volatile than normal with it being a 3 day weekend.
Oklahomahawk
Boras' Client
like a miracle
Is that the work of Trump's sharpie? Asking for a friend...
nsacpi
Expects Yuge Games
Oklahomahawk
Boras' Client
Well that IS what the J is for you know
nsacpi
Expects Yuge Games
only a chosen few spell it with a j
Runnin
Well-known member
Who are you talking about??? People on this board? Fantasy leftist devils placed in your mind by Steve Bannon?
http://www.enkiops.org/blog/2020/04...c5cSIA5EOdgz7ehCO8_o5_gMsdB3ek_G6Uf5JqgVke0a4
I've used this guy for hurricane tracking models for the last 5 years. He's usually right on the money. First time I've checked in on him discussing data about Georgia and COVID-19.
I've used this guy for hurricane tracking models for the last 5 years. He's usually right on the money. First time I've checked in on him discussing data about Georgia and COVID-19.
Oklahomahawk
Boras' Client
Or perhaps, just one...
nsacpi
Expects Yuge Games
usually there is only one great talent per generation...maybe