Share this page!

G Raghuram Profile picture
Twitter logo
1 May, 5 tweets, 2 min read
#COVID19India

Quite frankly, I don't (and couldn't know). I've mucked around from simple to ensemble #ML models over the last year, only to realize that something this complex cannot be modelled or predicted.

However, you can pick up what's happening, currently, based...
+
... off the standard epidemiological metrics - changes in trends, TPR, testing rates, recoveries etc. And what that means in terms of the near and medium-term horizons.

The most sensible benchmark would be TPR declining or levelling off against testing rates that are rising...
+
... against/ahead of case growth rates. If the data is passably reliable, it's a good indicator that a peak is to be expected, followed by a decline.

This would also show in R[t] hitting 1.0 on a decline and continuing to go under. If testing is doubtful, TPR is rising ...
+
... cumulative CFR is declining (temporarily) from an asymptotic trend, you don't know what's going on and certainly should not expect a real peak in infection in the region of interest.

Wish I could be more optimistic about this, but I don't see a quick turnaround ...
+
... regardless of what the daily case numbers suggest.

• • •

Missing some Tweet in this thread? You can try to force a refresh
　

Keep Current with G Raghuram

G Raghuram Profile picture

Stay in touch and get notified when new unrolls are available from this author!

Read all threads

This Thread may be Removed Anytime!

PDF

Twitter may remove this content at anytime! Save it as PDF for later use!

Try unrolling a thread yourself!

how to unroll video
  1. Follow @ThreadReaderApp to mention us!

  2. From a Twitter thread mention us with a keyword "unroll"
@threadreaderapp unroll

Practice here first or read more on our help page!

More from @UntergrundmannG

G Raghuram Profile picture
2 May
There are certain matters on which - for the survival (or a better future) of the collective - everyone (aka those whose survival is not threatened) contributes to or invests in. Education, Public Health, National Security are three well understood cases.

Those who ...+
... argue stupidly for "free markets" in these contexts haven't , quite clearly, thought through any of this and so end up betraying their immature understanding of how markets and societies generally function.

It can also be considered a form of insurance by/for those...+
...that have the most to lose (wealth, status, authority etc). There aren't too many of them - if you get the drift. So when things get bad, if it were everyone for himself, there's nothing preventing the larger # of threatened members of the society from banding together...+
Read 5 tweets
G Raghuram Profile picture
1 May
#COVID19India

I suffered brain damage late last night, having read the original pre-print. Read this simple explainer by @MenonBioPhysics instead.

The "model" is a bit like an astrologer's prediction. When the 'Mangal Muhurat' doesn't arrive, he says it's some Rahu-Ketu...
+
... and shifts his prediction and says something else. And when that fails as well, then he adds some other variable which has behaved unpredictably. You get the drift, if you're an Indian.

The problem though, is NOT that they're not epidemiologists. The issue is that ... +
... they're #clueless hacks who know one thing (in a bounded domain) possibly well. But cannot distinguish an extremely complex phenomenon, with fractal behavior/cascaded (or multi-scale) consequences and feedback loops, which is modelling-resistant.

It could be worse ... +
Read 4 tweets
G Raghuram Profile picture
1 May
#COVID19India: Summary View

Large TPR rises in most states, with the exception of AP, CT, MH and MP where it's negative.

Average TPR rises above 17%, in the range 3% to 34%. Most states between 11-22%
+ ImageImageImage
#COVID19India: Summary View

MH is the sole state with a negative weekly case growth.

CT, DL, MP and UP seem to have achieved the case doubling rate of 50 Days or more. However, this can only be validated by growth in testing.
+ ImageImage
#COVID19India: Summary View

AP maintaining its boost of testing growth. CT, MH and MP register testing growth higher than case growth.

Testing growth remains negative in BR, DL, GJ and TG. UP continues to have the least rise in testing. ImageImage
Read 4 tweets
G Raghuram Profile picture
30 Apr
#COVID19India: Summary View

Average TPR for the states rises to 16%+, ranging from 3% to 33%

MH, AP, CT and MP register negative TPR growth over the last week. Average TPR growth is 30% ranging from 13% to 116% for the states where it's rising.
+ ImageImageImage
#COVID19India: Summary View

MH, CT, DL and MP register a case doubling rate greater than 50 days viz. containment. This has to be seen against growth in testing accompanied by a decline in TPR.

Average weekly case growth for the other states ~ 40%, ranging between 28% to 98%.
+ ImageImage
#COVID19India: Summary View

AP, CT, MH and MP register testing growths > case growth for the past seven days.

AP persisting with the ramp up of testing.

Note the states with negative test growth - BR, DL, GJ, JH, TG. UP is the least positive. Correlate with their latest R[t]. ImageImage
Read 4 tweets
G Raghuram Profile picture
29 Apr
#COVID19India

This is another important read from @nature that highlights the problems with and limitations of R[t] and makes a case against over-reliance on it, especially for public policy wrt *easing* restrictions and return to 'normalcy'.
+
nature.com/articles/d4158…
The important takeaway is to understand the asymmetry in decision-making against the uncertainty implicit in this metric (aka what we cannot know).

If R[t] is rising above 0.9, should we enact restrictions? - is a VERY DIFFERENT Q, from the PoV of risk - than ... +
- Should we ease up restrictions because R[t] has declined from 1.3 to 0.8 (say) and seems to be going lower?

The first Q is a definite YES, with no doubts whatsoever. The second, however, is wide open, because things can reverse and get worse. So, you look at other ... +
Read 4 tweets
G Raghuram Profile picture
29 Apr
#COVID19India: Summary View

Latest TPR for most states has increased. The inference is that the spread continues unchecked and/or testing has been reduced. Average TPR rises to 16%.

MH, while lowest, registers a mild rise in TPR over the last week. CT is the sole negative.
+ ImageImageImage
#COVID19India: Summary View

Case growth for MH over the week is NIL aka no change. CT and DL both have doubling rates > 50 days and have the lowest case growths.

For the rest, it ranges from 30-130% WoW with an average of 46%
+ ImageImage
#COVID19India: Summary View

AP sees a boost in tests surpassing case growth. MH is low, but consistent with marginally negative case growth.

BR, DL, TG and UP all register negative testing growth aka reduction in testing levels.
+ ImageImage
Read 4 tweets

Did Thread Reader help you today?

Support us! We are indie developers!

This site is made by just two indie developers on a laptop doing marketing, support and development! Read more about the story.

Become a Premium Member ($3/month or $30/year) and get exclusive features!

Become Premium

Too expensive? Make a small donation by buying us coffee ($5) or help with server cost ($10)

Donate via Paypal Become our Patreon

Thank you for your support!

Follow Us on Twitter!

Like this author's thread?

Get emails when @UntergrundmannG has new unrolls!

Check out other threads from @UntergrundmannG!

Read all threads by @UntergrundmannG