@PTI_News
Well, this is news to me! As Deputy Director of IIT Kanpur, I am not aware of any such public protest by students. Maybe some students protested in their hostel wing!
This article appeared today in Print: . The reporter had called me in the afternoon and asked some questions about our model. I explained to her the conclusions including the fact that a large number of infected people in India showed few symptoms. 1/n
This is an appreciation tweet for
@covid19indiaorg
who have been diligently collecting Covid data for India at district level and have provided easy ways to access it. Without their portal, our work would not have been possible. Thank you! 🙏
@stellensatz
This thread summarizes key findings of our report .
What is the UP Model? At its heart is the decision not to shut down the economy during the second wave. This helps weaker sections of society as they are the worst sufferers of a lockdown.
<Update on 4/5>
@stellensatz
India curve continues in sync with predictions. If it does not suddenly decide to deviate, peak will be either today or tomorrow. Received an excellent suggestion to pin the latest India curve tweet, at least for next few days.
Somewhat amusing to see that most attacks on our report are ad-hominem. Some say we are validating use of ivermectin (we are not), some say we are only relying on government data (we are not), and some others say what about dead bodies floating in Ganga (an old practice). 1/3
A clatification: these predictions are NOT from SUTRA model, proposed by
@stellensatz
, Gen Kanitkar, and me. It is from a different group in IIT Kanpur. Predictions for third wave thru SUTRA model are under preparation and will be shared soon.
Nowhere in the report is the original point made that the virus seems to be affecting India less severely than many other countries (multiple reasons for this have been speculated). Anyway, that is what is reporting these days. I will be more careful with reporters in future. 4/n
<SUTRA's analysis of third wave>
@stellensatz
@Ashutos61
@Sandeep_1966
@shekhar_mande
It took us a while to do the analysis for three reasons. First, loss of immunity in recovered population. Second, vaccination induced immunity. Each of these two need to be estimated for future.
In Delhi it is around 1:25 while in UP/Bihar it climbs up to around 1:300. Unfortunately, the report has spun this in a different direction. The unreported cases become "missing ones". The implication being that our system has somehow goofed up by not detecting them. 3 /n
An excellent analysis of deaths in Wave-1 for India. Much better than - I do not like to use strong words but making an exception this time since this is best description - trash peddled in NYT article.
Starting a new thread on analysis of lockdown in various states. First, let us examine UP. The plot for UP and SUTRA projections for it from 1st March are below.
For those interested, we have not done any survey for our results. They are purely based on a mathematical model run on publicly available data. Although the model has been predicting the trajectory for the pandemic very well for past 3-4 months, it is still only a model. 5/n
To highlight this point, I mentioned that, according to our model, only one in 90 cases have been reported. And the unreported ones are primarily due to being asymptomatic. However, this ratio is not uniform across regions. 2/n
@stellensatz
@Ashutos61
@Sandeep_1966
@shekhar_mande
Starting a new thread for India. I updated India curve last on 14th with suggested peak at ~190K. Past few days have breached this value massively. This led to a discussion amongst us (me, Prof Sagar, and Gen Kanitkar).
Here are plots for the three scenarios. Blue curve is actual data. Orange one is model prediction until May. Dotted curves are three scenarios plotted from June.
I have published my rebuttal to
@sabya_economist
piece. I am specifically making the claim that the author‘s extraordinary claim that there is evidence of large scale voter deletion is not only backed by very flimsy statistical evidence, there is a possibility that he has
What is surprising is that the situation was brought under control so quickly. And the report explains how did that happen. I understand that many people oppose/support as per their ideology. This report is not meant for them, so they may kindly ignore. 😊 3/3
And conclusions of a model cannot be validated without cross checking it with actual data. One aspect of our model has been verified: its conclusions for pandemic trajectory have been confirmed by actual progression. 6/n
@ThePrintIndia
You missed stating that the article is in English, not German. 😂 All your responses are at the same level except the third point. And my quote was inserted by the journalist *after* I called her up and complained about the article. 🤦♂️
The other aspect, projection of total infections, can only be verified by sero-survey data. However, there have been few surveys, and even for the few that have taken place, there are doubts about the methodology adopted. So these conclusions remain unverified. 7/7
So it is not clear what will the final values be.
Peak timing: May 14-18 for active infections and May 4-8 for new infections.
Peak value: 38-48 lakhs for active infections and 3.4 to 4.4 lakhs for new infections.
<Update on 1/5>
@stellensatz
I have been exceptionally busy past few days, so not able to update as many plots as I would have liked to. In any case, India curve is the most anticipated one. And it is on track! Peak seems just around the corner!!
So the bottom line is: if there is no significantly faster spreading mutant, third wave will be a ripple. And if there is such a mutant, third wave will be comparable to first one.
However, if there is an immunity-escape mutant, all the above scenarios will be invalid!
@AbhishBanerj
These are just puff pieces written to make one feel virtuous. In reality, one of the authors of this piece has been literally begging for past several years to be hired even as contractual faculty at IITK. 😀
So lockdown came at the right time! BTW, the lockdown in UP is not very strict: many activities are permitted, including people moving from one place to another as long as crowding does not happen. It appears to be a very successful strategy.
7-day average of active infections reduced yesterday for the first time after 22nd Feb! An important milestone has been crossed as active infections determine load on health infrastructure. Better days are ahead now.😌
Finally, the website is ready. please check . Right now, it has plots of India, several states, and several districts of UP, Maharashtra, and Haryana. I will keep adding states and districts as they are simulated and stabilize.
Here is big news from
@IITKanpur
In an extraordinary gesture, our alumnus Mr Rakesh Gangwal, Co-Founder of IndiGo airlines has made one of the largest personal donations with a 100 crore contribution focused on supporting the School of Medical Sciences & Technology at IIT Kanpur
I have been asked by many people about details of the SUTRA model. We have a preprint uploaded at . It describes how we compute parameter values and phase changes.
Some focus on the two week period when situation was out of control and many died due to lack of bed or oxygen. Given the heath infra in the state and its population, it is no surprise that situation was out of control for a short period. 2/3
Another hard-hitting article by
@SubramanianKri
! He has been repeatedly calling out ideologically colored opinions on economic issues by many prominent economists including, tragically, his former advisor.
@stellensatz
More information on omicron variant. First, a more careful SUTRA simulation shows that contact rate (beta) jumped to 1 by Aug-end. It strongly suggests presence of a new mutant that was active by Aug.
<Update on 15/5>
@stellensatz
Another day of massive drop (by 11K+). One can now declare 8th May as peak of 7-day average of daily new infections at 3.92 Lakhs.
Recently various sources have pitched the Pfizer vaccine for India. This parlays the recent western ‘branding’ effort differentiating Pfizer, but it is known that this vaccine has some unique technical requirements.
This thread analyzes the Pfizer logistics in depth.
1/
Starting a new thread on countries. First UK since there is a lot of talk about another wave there. It does not appear to be a major one. The current phase is not fully stable, so the projection may change a bit, but not substantially.
@stellensatz
@GyanCMehta
The rapid spread of Omicron in China over the past one month has raised several questions:
1) Why is it happening in China now after such a long time despite vaccination? [1/18]
@ajitdatta
@AbhishBanerj
Excellent points
@AbhishBanerj
. There may be one more reason why many in media are not objective in their criticism: they know that if they become objective, they will have to acknowledge phenomenal work done by the govt in many important domains thereby helping BJP win more! 😀
Conclusions:
(1) Evidence so far suggests that there will be a mild third wave in India early next year.
(2) As observed during spread of delta, a mild lockdown (night curfew, restrictions on crowding) can bring down beta substantially. That will significantly reduce peak value.
<Update on 12/5>
@stellensatz
The drop in new infections continues! 7-day average plot is also showing a clear downturn. It appears that we peaked 8th May.
<Update on 8/5>
@stellensatz
Number came down yesterday. First day in this wave when seven day average has actually come down! It may not happen again today but we are not too far from the peak now.
@stellensatz
The assumption I made - India will behave similarly to SA - turns out to be wrong.
Indian trajectory is rising faster than projected earlier. We now have enough data to start doing preliminary predictions (as opposed to projections).
India is similar to South Africa in some crucial aspects. Both have young population (about 80% population is below 45 years). Both have high natural immunity (around 80%). Last wave in both was due to Delta-mutant.
This is a *much* better estimate of deaths in second wave. The conclusions are rather surprising!! 😳😳 Thank you
@Uncommo52204052
for brining sanity to discussions.
She called me twice when I was Deputy Director and showed similar disregard to civilized behavior. First time, I told her to give whatever she is saying in writing and then I will take action against her. She backed off then. 1/2
MP
@Manekagandhibjp
calls a man goonda, scoundrel, drunkard, cattle seller just cuz animal rights activist working for her claim so. Refuses to hear that the man hit dog for self protection, dog bit his daughter too. She threatens of molestation case if he doesn't remain quiet
@IshankModi
@PTI_News
Son, the march is supposed to happen *tomorow*. The PTI tweet suggests something has already happened. And as far as I know, time travel is not feasible.
Second, see the active infections plot below. It is on track to peak around 14th May. The jitter in new infections, therefore, does not appear concerning.
Finally India. Trajectory with data up to 11th indicates the peak on 23th Jan with nearly 7.2 lakh cases per day. Actual trajectory is already deviating significantly, and actual peak is unlikely to cross 4 lakh cases/day.
@stellesatz
Major update for website . It is now on https. Many districts added, and 20 states now. For Punjab, I am not yet confident about projected trajectory: phase is not stabilized yet. For WB, a new phase started on 2nd. Needs a few days data more.
One thing we took for granted was how Modi Govt managed the whole supply logistics during lockdown and pandemic restrictions. There was no shortage of food, basics even for poorest poor.
While Americans were queueing up and quarreling for toilet paper. My cousins state in US had
@TVMohandasPai
@IndiaToday
@rahulkanwal
Looks about right except for Karnataka. As per our model, India and Bihar are close to peak and UP, Delhi and Maharashtra have already peaked. We think that Karnataka is also likely to peak soon. So its R0 should be lower.
This article exhaustively collects order data and current delivery data for the US, India, EU/UK plus some other countries.
It specifically addresses the assertion that India failed by not making large bulk orders during mid 2020 like the west did by spending billions .
1/
There have been reports about significantly more deaths in UP (and elsewhere) than recorded. Nearly all of them are either anecdotal or use partial data. One needs CRS data for 2021, compare growth from 2020 against the trend of past decade to estimate excess deaths.
@TimesNow
keeps repeating misinformation that Prof.
@agrawalmanindra
, Lt. Gen. Madhuri Kanitkar and I are "government experts." We developed the SUTRA model on our own, not on behalf of GoI. Deeply disappointing to see Rahul S circulating this despite my clarifying it to him.
One of the winners of the Padma Shri this year is Dr Sanghamitra Bandyopadhyay. She is the first woman Director of the Indian Statistical Institute. Currently she is on the Prime Ministers' Science, Technology and Innovation Advisory Council.
Celebrate our scientists.
#science
Coming to India, we continue to have > 98% population with natural immunity. So I do not see any reason for concern. Over time, some percentage will lose natural immunity, and it may cause a small ripple, but it is VERY UNLIKELY that a significant rise will occur. [18/18]
<Update on 7/5>
@stellensatz
Another record-breaking day. Also, no of tests was lower than day before. The blue curve is clearly taking a different trajectory now. Is this a phase-shift? Will have to wait for a few more days to know for sure. Sorry to bring not-so-good news.😐
As articulated beautifully in this writeup, recent estimates for Covid-deaths during first wave have been ignoring basic principles of Statistics. 🤦 Therefore, it is hard to take them seriously.
Checking against these, one finds that new infections peak was within the range (May 4-8, 3.4-4.4 Lakhs), and active infections peak was slightly overestimated (May 14-18, 38-48 Lakhs). Not bad I suppose.🤨
When the curtains come down.
After keeping all of you updated with Covid19 data for the last 16 months, we will be stopping our operations from 31st October, 2021.
Why are we doing this? What will happen next?
This blog post has all the details 👇
We plot three scenarios:
Optimistic: vaccine immunity does not reduce
Intermediate: vaccine immunity is halved
Pessimistic: vaccine immunity goes completely
As can be seen, there is a mild third wave, with peak between 100-150K infections per day occurring sometime in Feb.
<Update on 10/5>
@stellensatz
Numbers were down significantly yesterday, but there is always a dip in numbers reported on Sunday (and even bigger dip on Monday). Still, the 7-day average has reduced a bit. Parameters continue to be stable. We should start going down soon.
<Update on 16/5>
@stellensatz
An even bigger drop yesterday (13K+)! Number of cases, as predicted, are dropping rapidly. The blue curve is following orange with a lag.
<Update on 6/5>
@stellensatz
I am finding it increasingly difficult to post updates. Hence, getting a website prepared that will do the job. Hope it will be ready by tomorrow. That will allow me to focus more on discussions.
Active infections stay on track. The blue curve is almost horizontal now. The load of health infra will start reducing once active infections start going down.
When we remove emotions from analysis, a picture closer to the truth emerges. The key point is that a phenomenon localized in time and space should not be taken to be representative of all time and space.
<Update on 22/4>
@stellensatz
India curves are moving in perfect sync for active infections. Peak value has gone up to ~37 lakhs which translates to ~3.7 lakh new infections. Peak timings remain the same.
How could 12,000 deaths go unreported in New York?
Perhaps NYT was busy pointing fingers at UP
If Western media wanted to “speak truth to power” in India, they crawled before power back home.
My piece for
@CNNnews18
@news18dotcom
Mitron, please read
@TimesNow
@Dr_AshokSeth
@VincentRK
Dear
@TimesNow
. Request you to please correct the nomenclature of Govt vs independent expert. We did not receive any funding or mandate from government to develop SUTRA. Thank you.
New infections peak about ten days before peak of active infections. Hence, the peak for new infections is expected during May 1-5. This was earlier projected to be between April 20-25. So we may have another two weeks of rise ahead of us.😐
@stellensatz
@Ashutos61
@Sandeep_1966
@shekhar_mande
This pandemic has a way of embarrassing those making predictions😀. We are indeed in the midst of second wave now. Of course, this wave is being driven primarily by Maharashtra. 1/n
Above also illustrates the pitfalls of modeling. One makes predictions based on imperfect information, and so has to be ready to accept failures... I will be tracking curve for India in this thread.