|
DontMockMySmock posted:Put a regression line on it  I couldn't find good data for the other factors (hospitalisation rate, etc.), but if someone else happens to have a source for it, I'll incorporate it in (perhaps as a colour gradient  )
|
#
?
Mar 19, 2020 20:13
|
|
|
|
| # ? May 5, 2024 22:17 |
|
|
Dr Sun Try posted:negative fatality rates????? That's when a woman gives birth while infected
|
|
#
?
Mar 19, 2020 20:14
|
|
|
morallyobjected posted:listen, I can only add so many dimensions, so if you figure out how to graph something in 4+ dimensions let me know: the bubbles' surface texture (shiny/rough) should relate to something maybe GDP cost?
|
|
#
?
Mar 19, 2020 20:28
|
|
|
morallyobjected posted:I got you fam
|
|
#
?
Mar 19, 2020 20:28
|
|
|
morallyobjected posted:
Chicken pox is highly leveraged and is driving the fit of your model. Use robust regression. Albino Squirrel posted:My take is that a) it's only due to insane luck and hard work that Ebola didn't become more widespread since it appears to have a similar R0 to COVID; b) holy gently caress smallpox was terrifying People with ebola aren't particulary contagious before they display symptoms. That makes it a lot easier to manage than some other diseases.
|
|
#
?
Mar 19, 2020 20:32
|
|
|
morallyobjected posted:
I feel like there could be a way to add another dimension by changing the data point to have more sides, e.g. from triangle -> square -> pentagon -> hexagon etc., to represent a spectrum. Edit: Excel might not support it, and no one should ever do it, but in theory it should be possible.
|
|
#
?
Mar 19, 2020 20:51
|
|
|
time for a blender charting plugin
|
|
#
?
Mar 19, 2020 20:54
|
|
|
Percentage of Opacity = Percentage of Cases Requiring Hospitalization.
|
|
#
?
Mar 19, 2020 20:58
|
|
|
Orientation of label in legend. I keep trying to type font and it autocorrects to don't every time.
|
|
#
?
Mar 19, 2020 21:04
|
|
|
Z axis, year of first appearance.
|
|
#
?
Mar 19, 2020 21:22
|
|
Yes, it's like a lava lamp.
|
ultrafilter posted:Chicken pox is highly leveraged and is driving the fit of your model. Use robust regression. here I made it a polynomial  I mean, here's the data table if anyone wants to go nuts with it: https://docs.google.com/spreadsheets/d/151k8recFTuAc_c6-C-I6DZkeKnTB6q8fFFKdqVcUpRs/edit?usp=sharing
|
|
#
?
Mar 19, 2020 21:26
|
|
|
It took me a bit to realize why they were so excited about jobs going down.
|
|
#
?
Mar 19, 2020 23:04
|
|
|
Alkydere posted:It took me a bit to realize why they were so excited about jobs going down. Yeah the decision to present the dates descending from left to right is a weird one.
|
|
#
?
Mar 19, 2020 23:06
|
|
|
morallyobjected posted:here I made it a polynomial 
|
|
#
?
Mar 19, 2020 23:46
|
|
|
Put both axes on log scales imo
|
|
#
?
Mar 19, 2020 23:47
|
|
|
vyelkin posted:Put both axes on log scales imo not imaginative enough. I did that AND I normalised the incubation periods and took the absolute value (it won't graph bubbles with negative values) so that the bubbles now represent how many standard deviations they are from the average incubation period of all of them, but you have no idea whether that's higher or lower. 
|
|
#
?
Mar 20, 2020 00:20
|
|
|
morallyobjected posted:here I made it a polynomial 
|
|
#
?
Mar 20, 2020 00:34
|
|
|
but i need the trend line for my report
|
|
#
?
Mar 20, 2020 00:34
|
|
|
taqueso posted:but i need the trend line for my report *sigh* my work is never done  also if anyone wants an even worse one, I made a version with error bars that, as far as I can tell, are probably meaningless: 
|
|
#
?
Mar 20, 2020 00:50
|
|
|
COVID is stored in the balls
|
|
#
?
Mar 20, 2020 01:01
|
|
|
thanks for the attempt but I needed a more accurate trendline
|
|
#
?
Mar 20, 2020 01:05
|
|
|
 Incubation is a color scale, obviously. But there's some important information that everyone else missed: size is based on when the disease emerged (as determined by a very brief wikipedia search.) Since I can't graph negative sizes to show BCE/CE, I decided the best way to do it is to normalize the size based on the beginning of recorded history, ~5000 years ago. I was too lazy to get it to write BCE, so enjoy the negative CE years values in the labels. This very important information makes a startling fact obvious: we can see that the center part of the graph is occupied only by smallpox, which is a very old disease. Based on the fact that there are no large (i.e., new) dots in the center of the graph, it must be impossible for any new diseases to emerge that have both infection rate greater than 3 and fatality rate greater than 10%. The graph shows it, so it must be true. This is how science works! EDIT:  More exciting science news! Not only did I fix the dates (it was rounding to the nearest 200 years because of an integer cast on the scaling), but when you add in a line chart in order of date, it shows a very clear straight-line path from MERS to COVID on the log-log chart. Much straighter than any historical period beforehand, at least when using this particular smoothing function. It's also very interesting that the rough log-Euclidean distance is about the same for every jump between diseases. That lets us predict what the next disease will be! Based on this, I have a message for the CDC: in roughly 2022, my model expects a disease with an infection rate of roughly 20 and a fatality rate of >1%. Start preparing now! Karia has a new favorite as of 01:50 on Mar 20, 2020 |
|
#
?
Mar 20, 2020 01:34
|
|
|
Karia posted:
taqueso posted:thanks for the attempt but I needed a more accurate trendline these are both excellent additions
|
|
#
?
Mar 20, 2020 02:22
|
|
|
Karia posted:
holy poo poo groundfloor  watching a nobel prize in the making
|
|
#
?
Mar 20, 2020 02:29
|
|
|
taqueso posted:holy poo poo groundfloor I want a co-author credit on the journal article
|
|
#
?
Mar 20, 2020 02:33
|
|
|
After further refinement, my model conclusively shows that the next anticipated disease will have a median infection rate of 25.78698 / case (95% confidence interval of 1.153992 to 567.2329) and a fatality rate of 0.196595% (95% confidence interval of 0.005191% to 7.445103%.) I expect an incubation period of -1.7 days (that means you start showing symptoms before you're exposed. Probably something psychosomatic. You know, like that episode of House MD where they're on the airplane.) It can be anticipated somewhere before 3233CE.  (The red stars show the 2 sigma confidence, since I have no idea how to draw an angled error bar in OxyPlot.)
|
|
#
?
Mar 20, 2020 03:37
|
|
|
|
|
#
?
Mar 20, 2020 03:53
|
|
|
wow eat poo poo nate bronze
|
|
#
?
Mar 20, 2020 04:07
|
|
|
Can we have a specially-jpeg'd version of this information for posting to Facebook?
|
|
#
?
Mar 20, 2020 04:08
|
|
|
Tree Bucket posted:Can we have a specially-jpeg'd version of this information for posting to Facebook? in general I'm all for enjoying bad graphs, but I feel like unless your Facebook friend list is very different than mine, the joke is going to be lost on people and it'll just end up being a bunch of misinformed people receiving what they think is legitimate information (because graphs are never wrong, and graphs with science-y words are DEFINITELY never wrong) and passing it on.
|
|
#
?
Mar 20, 2020 04:12
|
|
|
Can I get this in contour form?
|
|
#
?
Mar 20, 2020 04:32
|
|
|
Beartaco posted:Can I get this in contour form? Contour maps? Sure! Let's add the disease "badness" on there, calculated as (fatality rate * infection rate). I've also decided that the Fatality Rate scale should really be log base 2, rather than base 10, because obviously. The weird waviness comes from only evaluating the function at very course intervals on the log scale. I also manually set the interval spacing, because I'm pretty sure any reader would want all integer badnesses from 1-10.  Or did you mean as a vector file like an SVG? I can do that too, but I'm not sure I'm comfortable bumping this chart that far up the trustworthiness scale. morallyobjected posted:in general I'm all for enjoying bad graphs, but I feel like unless your Facebook friend list is very different than mine, the joke is going to be lost on people and it'll just end up being a bunch of misinformed people receiving what they think is legitimate information (because graphs are never wrong, and graphs with science-y words are DEFINITELY never wrong) and passing it on. Agreed. The initial versions, at least, are dumb, but a layman could absolutely think it was legitimate without thinking about it. Don't share unless you really know that people will get the satire. Karia has a new favorite as of 05:12 on Mar 20, 2020 |
|
#
?
Mar 20, 2020 05:10
|
|
|
I thought it was some kind of an astrological chart at first.
|
|
#
?
Mar 20, 2020 06:31
|
|
|
Karia posted:After further refinement, my model conclusively shows that the next anticipated disease will have a median infection rate of 25.78698 / case (95% confidence interval of 1.153992 to 567.2329) and a fatality rate of 0.196595% (95% confidence interval of 0.005191% to 7.445103%.) I expect an incubation period of -1.7 days (that means you start showing symptoms before you're exposed. Probably something psychosomatic. You know, like that episode of House MD where they're on the airplane.) It can be anticipated somewhere before 3233CE. Professor Karia, could this law also be used to determine when the first disease in human history occurred?
|
|
#
?
Mar 20, 2020 13:52
|
|
|
the awful charts are coming from inside the thread
|
|
#
?
Mar 20, 2020 15:25
|
|
|
Paladinus posted:I thought it was some kind of an astrological chart at first.
|
|
#
?
Mar 20, 2020 15:44
|
|
|
Awful/funny graphs and charts: Ebola in retrograde
|
|
#
?
Mar 20, 2020 16:37
|
|
|
Red Bones posted:Professor Karia, could this law also be used to determine when the first disease in human history occurred? Sure! Disease 0 likely emerged around 3000-3500 BCE, and had a median infection rate of 7.14 (95% confidence interval 6.1 to 8.4) and fatality rate of 750% (95% confidence interval 6.3 to 89000%.) That's far worse than Ebola, with a badness of 5300. For an epidemic that bad, there's only one possibility:  We need to rediscover this disease, use it to bring George Romero back from the dead, and then get him to direct a movie about this outbreak. I am seriously regretting my vow to do all of this in Oxyplot rather than just scribbling on the chart in Powerpoint or something, it takes forever.
|
|
#
?
Mar 20, 2020 17:58
|
|
|
Karia posted:Sure! Disease 0 likely emerged around 3000-3500 BCE, and had a median infection rate of 7.14 (95% confidence interval 6.1 to 8.4) and fatality rate of 750% (95% confidence interval 6.3 to 89000%.) That's far worse than Ebola, with a badness of 5300. For an epidemic that bad, there's only one possibility: Thank you! My god, a fatality rate of 750% really puts our current problems into perspective.
|
|
#
?
Mar 20, 2020 21:45
|
|
|
|
| # ? May 5, 2024 22:17 |
|
|
|
|
#
?
Mar 20, 2020 21:58
|
|
