Federico Andres Lois Profile picture
Sep 24, 2020 24 tweets 5 min read Read on X
The arrival of a vaccine in the context of COVID can be modeled using game theory as a gamble over the expectation of the final death toll. Most countries will have negative payoff after August 2020. Change my mind. cc @LDjaparidze
Let's make it more interesting... Do you agree?
Context is king. For a gamble to exist we need to define clearly the parameters to observe the likely expected result. Let's start with vaccine efficacy (VE). What do you think is the range most manufacturers are looking for?
I am not an expert, but I happen to talk with people that are on the business. They tell me for a respiratory virus a 55% target is quite good. But, let's figure out if that is true. This is Moderna protocol: modernatx.com/sites/default/… Image
But you know why we would trust one manufacturer when we can get another one. This is Pfizer protocol: pfe-pfizercom-d8-prod.s3.amazonaws.com/2020-09/C45910… Image
If we modeled such a vaccine then, would be fair to use VE of 77% then?
Interesting. I was going to go god mode, but now I am interested. Do you believe the gamble should consider the vaccine VE should be:
Ok, so you choose to go lower than 77% will go 70%. To please the other side, I will make one more assumption to equalize. You can apply the vaccine to a 100% population overnight.
Let's say we have a time machine and we can come back to March when mitigations started to happen all over the world. How long would you be willing to mitigate before coming back to normal life in order to protect the vulnerable? Real Normal Life!!! January.
We are all burned out by mitigations. I get it, and yet we keep doing it for our vulnerable. So as context is king. This is what our most likely scenario is for Madrid with data until July looks observing continuous mitigations and the expected summer behavior observed at London. Image
The big question is how that scenario looks like in terms of what is really happening. Fitted from data up to July 9th, as you can see for September it predicts roughly 22 deaths per day. Currently is around 30 per day. comunidad.madrid/sites/default/…
If our 70% vaccine would be applied tonight to 100% of the population after 199 days after the initiation of lockdown. Do you think it would be able to beat 90 days of Sweden fitted trajectory?
OK, I have all people with yellow and blue shirts here it seems. Given the Stockholm trajectory was estimated to have the following isolation parameters:
- Vulnerable: 0.95
- Healthy: 0.2
Stockholm looks like this: Image
With under 1 deaths per day on average. Does that looks endemic to you?
Do you think the Sweden strategy is optimal?
What I can say is this: The optimal strategy for Stockholm no matter when the vaccine comes for our model does not exist. But, if the vaccine does not arrive before December there is a better strategy:
Vulnerable: 0.95
Healthy: 0.16 Image
Does that looks like right there on the margin of error?
OK. For all uses and purposes, the Sweden strategy under our isolation model is optimal; unless of course you want to fix bad government policies. That's for another thread. Do you think the Madrid strategy would stand a chance against it?
The last graph of this series and the one that solves the gamble under our isolation model is quite complex. It is a level of abstraction on top of this linear thinking. We will respond to the following question:
If an oracle would tell me when the vaccine will be available. Is there an optimal strategy that would minimize the death toll given that date?
But, it doesn't end there. At the same time is good to know how other usual strategies found in the wild would fare under the model.
Isolations:
- Very strict: 0.9 (Sweden on vulnerable)
- Strict: 0.65 (Lockdown on average, Masks)
- Moderate: 0.4 (No Schools, Masks)
- Almost normal life: 0.13 (Schools, No Masks, No stadiums [???]).
QED. With very general assumptions (remember I wanted to go god-mode), after August on Madrid even if our magik vaccine comes; under our epidemic model from the point of view of game theory the expectation is that it will be a losing proposition. Image
Do you want to know if it is possible in 90 days to correct government policy mismanagement?

• • •

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

Keep Current with Federico Andres Lois

Federico Andres Lois 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 @federicolois

Apr 24
1/ After almost 1.5 years of studying cancer research for personal reasons, I arrived at a realization that prompted me to write this tweet. I will lay out the hypothesis in this thread.
2/ Disclaimer: I am not a formally trained health researcher. More like a very curious and tenacious guy with a 15+ year background in research, development, & reproducibility in computer science (computer science).
3/ I am putting the hypothesis out there because it may make sense to others doing field work. Feel free to dissect this hypothesis, find holes in it, and play devil's advocate. We will all come out smarter from it.
Read 33 tweets
Jan 3
1/ There is a very perverse dynamic on how Chavism (aka "the communist socialism") works. Let's use Argentina as the example. Over the first 20 years they initiate a process that we could call "Earnings Substitution" that will seal your fate over time.
2/ Your earnings/salary is going down and at the same time "subsidies" start to go up in order to fool people into think that nothing has changed. This works because the dirty job is done by inflation which is a much slower process.
3/ By the time people starts to realize that something is wrong, because some critical goods are not available (medicine, food, you name it) or inflation enters a death spiral; most people already depend on subsidies for spending.
Read 6 tweets
Aug 2, 2023
1/ Recently some interesting papers have been doing the rounds in the health community. To me the most interesting ones have been the GlyNAC paper and the more recent Taurine deficiency as a driver of aging papers.
2/ Disclaimer: While I have been researching this for a year and even executed an experimental protocol tailored for myself based on the GlyNAC paper, I am NOT a health professional, and I am just taking my health into my own hands. This is not advice of any kind.
3/ Disclaimers aside, why do I think these 2 papers are interesting? First because the claim (if true) is a game changer. And second because they may be related but I haven’t seen this relationship spotlighted by anyone.
Read 14 tweets
Apr 1, 2023
This just confirmed the weaponization of block lists. If enough people/bots block and mute you, they are essentially cancelling you. I find lots of people with I have never interacted with that has me blocked. Assuming there are third party block lists and block networks.
Normally that is an issue in general. Anyone that has done reinforcement learning had figure out (usually in the worst way) that you have to be incredible cautious with penalties. They are very prone to be gamed.
Read 4 tweets
Mar 20, 2023
1/ I asked #GPT4 to review our paper with @LDjaparidze. This is what happened and what I learned in the process. medrxiv.org/content/10.110…
2/ Since the general problem that practitioners find (in the worst way) is always training set tainting (guilty-as-charged). Habits die hard, the first thing I did is asking to do a review of the paper without any extra knowledge about what the paper says
3/ From the response alone I learned 2 things. First, our paper title was deadly accurate. I also learned that it has no information whatsoever on it, as the entire response can be generated from understanding the title itself.
Read 10 tweets
Feb 12, 2023
1/ I found this paper intriguing so my first step was to verify you can trigger this behavior on ChatGPT. It is actually pretty easy.
2/ Since I am doing it by hand I started with a very simple prompt.
3/ I have been arguing that this trying to constrain the model is actually harming it before. This is one of those cases. The good thing is that at least for you just add "Use the tokens" at the end of the request when it refuses and it will do it properly
Read 5 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

Don't want to be a Premium member but still want to support us?

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

Donate via Paypal

Or Donate anonymously using crypto!

Ethereum

0xfe58350B80634f60Fa6Dc149a72b4DFbc17D341E copy

Bitcoin

3ATGMxNzCUFzxpMCHL5sWSt4DVtS8UqXpi copy

Thank you for your support!

Follow Us!

:(