A little fact checking goes a long way. There’s enough fake news as it is. As you yourself admit that it’s hearsay, what good could possibly come of spreading false information? 
5 Likes
part of it is hearsay, part of it is on the news. but unfortunately most of the material is in chinese so even if i post it here you guys can’t read it.
what is certain is: in its 70 year history china had never locked up a whole city, shut down factories for weeks, and admitted so many died.
that is a sign of how seriously china is taking it.
4 Likes
another reason why this is worse than flu is the permanent damage it does to your lung.
you can recover from a flu fine, but with this virus your lung never recover.
I can read Chinese. I wouldn’t mind seeing if you’d like to share. And others who don’t speak Chinese can translate it if they’re interested.
3 Likes
Is there any evidence of anyone initially healthy and fit suffering permanent lung damage?
That sounds like the effect of pneumonia, which older people can fall to after corona virus?
I’m not in panic, but I follow the news on covid-19 closely. I am very sure I will catch it at some point, and as a parent of a toddler who’s in kindergarten, I can tell you I’ve been having colds constantly for the last 10 weeks or so. 
Not exaggerating. So, I don’t see a realistic chance for me to avoid any virus that finds its way to either the kindergarten, or to my partner who travels a lot for work, or to other family members. I’m not worried for myself, but I worry a little bit for my parents.
3 Likes
Within the previous days, I read a few very interesting thoughts about covid-19’s spread.
One was an interview with a virologist from Berlin’s Charité hospital, who said he expects that, in the end, 60 to 70% of the global population will catch the virus, but it matters a lot in which pace this happens. If all countries manage to slow the spreading down, it will maybe take two or three years for the virus to spread, and then there’s time in the summer, when the virus will probably not be able to spread quickly, to prepare for the winter and improve the health care system in the meantime (he was especially talking about Germany here). If not, then it will spread exponentially and our hospitals are not up to care for all the patients.
So, this should imho be a good reason for everyone to be cautious and try not to become a spreader, even if you’re not in the risk group.
6 Likes
Actually its graph of found infected, not graph of speed of spread, so it useless.
But still interesting that they look nearly same.
Now there are 4,335 in South Korea
And 47,163 in Hubei 14 days later
1 Like
Looking to the numbers (as usual for me): here you can find some more data:
https://gisanddata.maps.arcgis.com/apps/opsdashboard/index.html#/bda7594740fd40299423467b48e9ecf6
if you enlarge the portlet with the plot, and you select logarithmic tab view, it appear clear to me that there is a big acceleration of the reported cases. In fact, if you have an exponential trend in the number of reported cases, you should have a stright line in a log scale. Actually from the graph you can see that the trend of registered case outside China raise more than a stright line. This suggest to me that the acceleration is more than exponential in this phase.
Note: those are official numbers reported from different government collected by Johns Hopkins Institute.
See at yellow curve:
3 Likes
If you’re interested in modelling a generic infection spread rate, the SIR method is very straightforward and explained well in the link.
As you suspect, it’s not a simple algebraic equation.
Also, please note that the JH Gis tracker only does government reported infections, which is not very relaible. There is a paper here that uses some reasonable assumptions and places the actual number of infected much much higher than govts reports.
2 Likes
I don’t see this… Also, if it was growing superexponentially, that would be very weird; not only the number of infected would then be growing, also the infection chance of someone with the virus would be increasing.
2 Likes
Compare line AB with CD.
As this is a log scale graph, any function of the form x^n will have a straight line. Anything that grows faster than a straight line on a log must therefore grow faster than an exponential.
Therefore, CD represents exponential growth while AB is faster than exponential.
2 Likes
Uhm, no, a straight line is any straight line, it can have a nonzero slope and still be exponential.
I’d say AB is a pretty straight line, to be honest… Of course data is not perfect, so it curves up very very slightly, but not a significant amount.
Also, functions of the form x^n will have downward curving lines (that is, their slope will decrease on a log scale): they are subexponential. Functions of the form a^x will have straight lines, as exponential functions (assuming x is your variable). The variable a decides the slope of the line.
3 Likes
They both roughly look like exponential growth, but at different rates. The CD line looks almost flat, but I believe it does have some small positive slope.
On a semi-log plot (X-axis is linear, Y-axis is logarithmic), any linear trend upward represents exponential growth. The slope of the line is related to the doubling rate.
On a semi-log plot, any growing polynomial function would have a curved (concave) shape to it. For example: x^3 in semi-log plot
3 Likes
My mistake.
Also, I just read some following posts and I never meant to imply that the linear interpolation itself is the proof, but that the data points for AB are seemingly curved as opposed to CD. As said later, one looks concave and the other doesn’t.
1 Like
Guys, it wasn’t my intention to open a mathematical dispute on the data.
The Y-axis is a base 10 log scale. if the derivative first raises with the increase of X (time) it means that the curve is not a straight line but a curve with an upward concavity (second derivative > 0).
This means that the plot grows faster than exponential. That’s it.
But it is just a curiosity. Don’t change anything in terms of understanding what is happening, I think.
2 Likes
First case reported from my region of England, which has about two million people. OX (skull and crossbones emoji – might come in useful ^^)
1 Like
24 hospitalized in Moscow because some person had it returning from abroad.
My apologies, but it’s almost impossible for me to not get into mathematical disputes when a chance arises… It’s a consequence of having studied mathematics
A convex curve on a logarithmic graph implies that the data grows faster than exponential. However, the graph you showed, shows only very slight convexity, most of the curves are more or less straight, or even significantly concave.
3 Likes