but, we might have the most blatantly dishonest government in the oecd, right now.
Friday, May 15, 2020
we appear to be reopening slowly in ontario, but i don't think we're even closing to peaking here yet.
so, i will expect the numbers to climb rather dramatically, even if they end up distorted by bad reporting. it's not clear how it's going to get reported, and i think i've decided to largely just ignore the government rather than pretend i can analyze fraudulent data.
but, it's not going to climb because of reopening, even if the government tries to argue the point. we just haven't come close to peaking yet.
so, i will expect the numbers to climb rather dramatically, even if they end up distorted by bad reporting. it's not clear how it's going to get reported, and i think i've decided to largely just ignore the government rather than pretend i can analyze fraudulent data.
but, it's not going to climb because of reopening, even if the government tries to argue the point. we just haven't come close to peaking yet.
what that means is that if you were to decompose the canadian curve into different provinces, you'd see something more like the peaks and decays in the european countries.
and, likewise, if you were to decompose the american curve into different states, you'd see something more like the peaks and decays in the european countries, too.
if you want to test this question of the efficacy of social distancing, you need to look more locally. so, what's happening in georgia, specifically? but, be careful - did georgia peak before it reopened?
and, likewise, if you were to decompose the american curve into different states, you'd see something more like the peaks and decays in the european countries, too.
if you want to test this question of the efficacy of social distancing, you need to look more locally. so, what's happening in georgia, specifically? but, be careful - did georgia peak before it reopened?
i got that picture from here:
https://www.motherjones.com/kevin-drum/2020/05/coronavirus-growth-in-western-countries-may-14-update/
kevin drum is arguing there that he expects that the united states should see a peak and a rise due to the loosening of social distancing, and that if we don't see such a thing then the epidemiologists will have hell to answer to.
i've been arguing that social distancing is pseudo-science from the start. but, i actually don't expect the curve in the united states to fall off quickly - and don't expect the curve in canada to, either. the reason is that canada and the united states are both very big countries with several epidemics, whereas a country like sweden is really just one big city. you'd be better off being as specific as possible, but you could fudge a comparison of all of italy to new york city; comparing individual countries in europe to the entire united states is pointless.
rather, you'd want to compare canada or the united states to the european union, and it's easy to see intuitively that if you superimpose these graphs on top of each other then they will intersect to form a plateau.
so, i expect that the united states will, indeed, plateau, but it was has less to do with social distancing restrictions being lifted and more to do with the fact that you're dealing with 50 different curves pasted on top of each other, rather than just one.
for that reason, the flattening curve in canada is also probably predictive. it's not going to fall like it did in italy, it's going to take a while...
https://www.motherjones.com/kevin-drum/2020/05/coronavirus-growth-in-western-countries-may-14-update/
kevin drum is arguing there that he expects that the united states should see a peak and a rise due to the loosening of social distancing, and that if we don't see such a thing then the epidemiologists will have hell to answer to.
i've been arguing that social distancing is pseudo-science from the start. but, i actually don't expect the curve in the united states to fall off quickly - and don't expect the curve in canada to, either. the reason is that canada and the united states are both very big countries with several epidemics, whereas a country like sweden is really just one big city. you'd be better off being as specific as possible, but you could fudge a comparison of all of italy to new york city; comparing individual countries in europe to the entire united states is pointless.
rather, you'd want to compare canada or the united states to the european union, and it's easy to see intuitively that if you superimpose these graphs on top of each other then they will intersect to form a plateau.
so, i expect that the united states will, indeed, plateau, but it was has less to do with social distancing restrictions being lifted and more to do with the fact that you're dealing with 50 different curves pasted on top of each other, rather than just one.
for that reason, the flattening curve in canada is also probably predictive. it's not going to fall like it did in italy, it's going to take a while...
this graph kind of enforces what i've been saying about canada just being last to get hit....and i don't do this very often because i kind of don't need to. graphs are just tools, remember - they help you understand, but they're only as useful as they are necessary.
"do i need to draw you a picture?"
most of the time, i don't need to draw myself pictures, so i don't bother - i can get the point by looking at the information in the chart, and i'll just tell you about it. you can draw your own fucking picture, if you have to.
but, somebody drew this graph for me, and what does it say?
it says canada is peaking last - which has been obvious for a while.
and, why is it peaking last?
because it's furthest away.
that is all.
really.
"do i need to draw you a picture?"
most of the time, i don't need to draw myself pictures, so i don't bother - i can get the point by looking at the information in the chart, and i'll just tell you about it. you can draw your own fucking picture, if you have to.
but, somebody drew this graph for me, and what does it say?
it says canada is peaking last - which has been obvious for a while.
and, why is it peaking last?
because it's furthest away.
that is all.
really.
is montreal peaking?
well, they're at a little below 2200 deaths, and there is some suggestion that it's a little undercounted.
this was my projection:
3) montreal: (2800, 8399, 13999) [projection: low end]
the first number is a low estimate, the second is a middle estimate and the third is a high estimate. these are dependent not on the efficacy of social distancing but on different estimates of the mortality rate.
if they're actually at 2500 or so, they should cross the minimum barrier within a few days, meaning the numbers should start to slow down quite a bit.
remember: i don't think that total excess mortality much of anywhere is going to have much to do with what the government has or hasn't ordered people to do. so, i don't think that the end results are very dependent on government policy.
my low numbers have been a little too low in most places, but almost nowhere has actually reached the middle number yet. the only exception is new york city, where the population density is so dramatic, and i appear to have underestimated the effects of diabetes.
so, a reasonable expectation in montreal in the end is somewhere around 4000, regardless of what the government does.
well, they're at a little below 2200 deaths, and there is some suggestion that it's a little undercounted.
this was my projection:
3) montreal: (2800, 8399, 13999) [projection: low end]
the first number is a low estimate, the second is a middle estimate and the third is a high estimate. these are dependent not on the efficacy of social distancing but on different estimates of the mortality rate.
if they're actually at 2500 or so, they should cross the minimum barrier within a few days, meaning the numbers should start to slow down quite a bit.
remember: i don't think that total excess mortality much of anywhere is going to have much to do with what the government has or hasn't ordered people to do. so, i don't think that the end results are very dependent on government policy.
my low numbers have been a little too low in most places, but almost nowhere has actually reached the middle number yet. the only exception is new york city, where the population density is so dramatic, and i appear to have underestimated the effects of diabetes.
so, a reasonable expectation in montreal in the end is somewhere around 4000, regardless of what the government does.
so, i got a first pass finished this morning, but it's so dry in here that i'm literally bleeding. i'm hoping the temperature outside has stabilized from that freak cold snap, so that i can begin the process of bringing the humidity levels inside up to something more normal. and, i think i'm feeling better already, but let's hope it's not fleeting.
so, i guess i blew the whole week, but let's hope that things renormalize themselves a bit better from here on in.
i'm going to stop to do some cleaning and just generally run the hot water for the next several hours.
i still need to rebuild the appspot portion of the site into the blogger front-end, because all of the files that i built up at the beginning of the month seem to have strangely disappeared. there's also going to be a cross-referencing phase.
but, the bulk of it is done, and i would expect to be able to publish it by monday.
i think there's a cop upstairs, but the property owner is technically not home and hasn't been for months. so, let's hope we can avoid the air conditioner for most of the year...
so, i guess i blew the whole week, but let's hope that things renormalize themselves a bit better from here on in.
i'm going to stop to do some cleaning and just generally run the hot water for the next several hours.
i still need to rebuild the appspot portion of the site into the blogger front-end, because all of the files that i built up at the beginning of the month seem to have strangely disappeared. there's also going to be a cross-referencing phase.
but, the bulk of it is done, and i would expect to be able to publish it by monday.
i think there's a cop upstairs, but the property owner is technically not home and hasn't been for months. so, let's hope we can avoid the air conditioner for most of the year...
Subscribe to:
Posts (Atom)