Forums  > Off-Topic  > COVID19  
     
Page 4 of 5Goto to page: 1, 2, 3, [4], 5 Prev Next
Display using:  

TSWP


Total Posts: 448
Joined: May 2012
 
Posted: 2020-04-03 10:36
Nikol, I don't disagree with you, your analysis is excellent...

I am just trying to say that if there is a core problem, the root of it all that puts things in motion or stops them, is the knowledge that the virus can be killed, because that knowledge allows politicians to act and let people go back to work and return the economy to its normal functioning.

Politicians are in panic mode now, they don't know what is going on and they have decided to shut everything down, typical knee-jerk reaction, typical politics...

This is not an economical crisis, not a medical crisis either, this is purely a political crisis where we see fully the extent of uselessness of politicians as decision makers (nothing new under the sun...).

So the key solution to this crisis is for politicians to have in their hands what they need to restore the previous status quo: a vaccine that guarantees to avoid pandemic contagion. Once that is available, politicians will re-open the gates.

Anticipated partial re-opening are possible, but the risks are that the pandemic will come back and so I am not sure politicians will be willing to take that kind of risk, maybe they'll do it once to try to restore normality, and if it does not work they'll shut down everything again...

gmetric_Flow


Total Posts: 39
Joined: Oct 2016
 
Posted: 2020-04-03 14:09
@TSWP
Agreed, all political, fiscal, and monetary action is at best palliative. Whereas a vaccine is an absolute solution, an effective anti-viral therapeutic (although not prophylactic) would be a major step forward, and probably "good enough" for reopening.

TSWP


Total Posts: 448
Joined: May 2012
 
Posted: 2020-04-07 07:10
Yeah, agreed, vaccine or any medicine that heals people when sick... that would do...

nikol


Total Posts: 1172
Joined: Jun 2005
 
Posted: 2020-04-07 18:07
From skeptical I am turning into cynical being

https://www.businessinsider.nl/inovio-coronavirus-vaccine-trial-starts-in-philadelphia-kansas-city-2020-4/

"Inovio said it was scaling up its production capabilities and aiming to have 1 million doses available by the end of 2020, which could be used in additional clinical trials or for emergencies"

I would believe into humanity if this kind of material would be shared immediately. There is a success story about polio vaccine cooperation and use (USSR to allies and to Japan) in 50-60s.

https://en.wikipedia.org/wiki/Cold_War_tensions_and_the_polio_vaccine

chiral3
Founding Member

Total Posts: 5163
Joined: Mar 2004
 
Posted: 2020-04-15 02:12
I was just thinking how many big names in science and math left us in the last year and then I got a note that COVID19 got John Conway.

Nonius is Satoshi Nakamoto. 物の哀れ

pj


Total Posts: 3530
Joined: Jun 2004
 
Posted: 2020-04-15 12:36
RIP John Horton Conway

On a lighter note.
The Sun article on Covid-19
It's Nicol's fault that I went there!

The older I grow, the more I distrust the familiar doctrine that age brings wisdom Henry L. Mencken

nikol


Total Posts: 1172
Joined: Jun 2005
 
Posted: 2020-04-15 14:03
At my school time I learned J.Conway's Life game. It impressed me so much that I have spent perhaps 1 month of my after school time rigorously testing various configurations on the check paper.
RIP.

@pj, sorry for SUN, but its readers outnumber academicians in voting rights. So, the result.

svisstack


Total Posts: 355
Joined: Feb 2014
 
Posted: 2020-04-28 23:06
https://www.bloomberg.com/news/articles/2020-04-27/wall-street-quants-are-turning-their-skills-to-the-virus-fight

First Commander of the USS Enterprise

nikol


Total Posts: 1172
Joined: Jun 2005
 
Posted: 2020-04-30 10:22
Simulator by MIT. Methodology is linked there

https://analytics-tools.shinyapps.io/covid19simulator/

frolloos


Total Posts: 124
Joined: Dec 2007
 
Posted: 2020-05-21 18:29
I am not sure it is a good thing Wall Street quants (assuming Kolanovic is considered a quant by the community) are turning to Covid-19:

https://twitter.com/carlquintanilla/status/1263170459450978314


The tweet is amusing, the chart is not even wrong.

No vanna, no cry

chiral3
Founding Member

Total Posts: 5163
Joined: Mar 2004
 
Posted: 2020-05-21 18:49
I saw that data this morning. Not sure I knew what to make of it but didn't think too hard about it.

Nonius is Satoshi Nakamoto. 物の哀れ

NIP247


Total Posts: 550
Joined: Feb 2005
 
Posted: 2020-05-22 20:27
While I may agree with the conclusions, I don’t like that MK is mixing up R0 and R ( or Re ), nevermind that Re is apparently so difficult to estimate that it is not used for management of the crisis ( but it’s a number that gives the illusion of understanding...)

On your straddle, done on the puts, working the calls...

chiral3
Founding Member

Total Posts: 5163
Joined: Mar 2004
 
Posted: 2020-05-22 23:51
Funny you should say that - that was the first thing that jumped out at me. I didn't say anything but, in my mind, I said that if a data scientist wants avoid giving ammo to the "I didn't know JPM hired epidemiologists" crowd, they'd get those details right.

I don't agree with the conclusions. 7 day lag (not sig) coupled with unknown etiology - behavior, flattening, etc. Nobody coming out of a lockdown in the last three weeks is behaving like they would 3 months ago. Bottom line is the data is shit. Cases are rising right now compared to the previous 2nd deriv in many states and places, only because there's increased testing. Data isn't standardized, etc.

Nonius is Satoshi Nakamoto. 物の哀れ

Maggette


Total Posts: 1251
Joined: Jun 2007
 
Posted: 2020-05-23 08:06
Is the "devastating piece" publicly available?

That's a piece of political propaganda, not "data science". It is one thing if the machine learning folks (which I am basically part of) try to contribute. But just from the twitter post I can infer the article is making statements that can't possibly made at present. Like chiral3 said: the data is scarce and messy, the problem is quite complex.

It is a bad rep for my profession if people think they ride in with some public data, some sci-kit learn library an "generate insights".

I worked in very different domains. Finance, commodity trading, logistics, automotive, aerospace.
The rules has always been:
1) look at data
2) shut the fuck up and listen
3) only break rule 2) to ask questions. Lot's of them. Then back to 2).
4) Try to make sense of what people with lots of domain knowledge and experience in the field tell you and what you learned from 1). Try to incorporate as much physics, engineering and domain know how in your way to interpret data.
5) Be very very careful with the conclusions you draw from data and even more careful to publish it as an result and even more so if the result doesn't natch what you learned in 2) and 3). It does happen, more often than not it is an artefact of your data/ the way you look at the problem.

In regards to my Covid experiments with public data....every "pattern" I think I have found kind of went away after I tried to adjust for the amount and density of testing using some bad proxy. I would never ever conclude that I can read ANYTHING from that data.

I didn't read the article but I also have a feeling there is a lot of "shootings are more likely to occur when police arrives on a scene with horns and lights than after they left. So let's not send them" going on.

We are in the world of uncertainty here...not risk.
We need more "real science" (like laboratory and shit) here to make sense of the data and we need much more data.

Ich kam hierher und sah dich und deine Leute lächeln, und sagte mir: Maggette, scheiss auf den small talk, lass lieber deine Fäuste sprechen...

NIP247


Total Posts: 550
Joined: Feb 2005
 
Posted: 2020-05-24 21:22
Maggette, it's only political because MK is a marketer at heart, not because of some wider political agenda. He knows what gets talked about and therefore maximise exposure. That has been true going back to the report on varswap hedgers affecting the closing price. Clients just loved that. It sounded like you unveiled some insider knowledge about how the market "really works". And he's been on a roll ever since. This has clearly worked out for him.

With regards to armchair epidemiology, it's true that anyone with excel or even a sci-kit learn library is toying around with the data and making predictions. However, this doesn't mean that the approaches used in finance, commodity trading, logistics etc are applicable either. There is way more uncertainty in this field, and it seems to me that it demands some completely different skill-set, a little like meteorology. And the data is inherently bad and difficult to compare or use in real time in the way other disciplines would approach modeling and decision making.

On your straddle, done on the puts, working the calls...

Maggette


Total Posts: 1251
Joined: Jun 2007
 
Posted: 2020-05-24 23:13
Agreed. I didn't want to imply that me beeing exposed to applied science and engineering in different domains somehow enables me to solve these problems. Just that you get a feel for limits of data driven methods applied to a few data points generated by a real world cyborg. Messy, instationary and most important not enough.

The models aren't bad. Armchair or Epedimiologust. As additional data points in a decission making process they are valuable...the mindless application of a single model to make decisions is kind of crazy.

I also agree that other decission making approaches or ensembles of these should be considered. Hence my distinction of risk vs uncertainty. Even if that leads to frustrating heuristics like maximin, minimax or hurwicz alpha and/or some bayesian approaches with wild priors.

Ich kam hierher und sah dich und deine Leute lächeln, und sagte mir: Maggette, scheiss auf den small talk, lass lieber deine Fäuste sprechen...

EspressoLover


Total Posts: 445
Joined: Jan 2015
 
Posted: 2020-05-25 00:15
One problem with simply looking before and after a change in lockdown orders, is that it doesn't seem like the timing or details of lockdown orders actually changed people's behavior that much. When you actually look at the cell phone tracking data, most people started staying home at some point in March. But there were no discontinuities on the actual date that local or national lockdown orders were put in place.

Facebook's tracking data seems to confirm a roughly similar result. We can also compare across cities, for example Stockholm to Nashville Apple's data show almost identical stay-at-home behavior. Despite the former having one of the most permissive lockdown orders in the Western world, and the latter one of the strictest. It seems pretty much like you have a population of risk-averse people who were going to stay-at-home even if the government didn't explicitly tell them to, and a population of risk-tolerant people who probably aren't scared of legal repercussions anyway. There's not that many people on the margin, who aren't worried about infection but fear getting arrested.

The point is that it doesn't seem plausible that lockdowns explain any more than a small fraction of the variance, either cross-sectional or time-series, in transmission rates. With that in mind, it's pretty implausible to just compare transmission rates between two points in time or space, and ascribe the differences to lockdown status. Almost certainly there are other factors (including just randomness) that overwhelm the impact of lockdown policy.

Good questions outrank easy answers. -Paul Samuelson

chiral3
Founding Member

Total Posts: 5163
Joined: Mar 2004
 
Posted: 2020-05-25 00:57
Thanks for the links EL.

Nonius is Satoshi Nakamoto. 物の哀れ

keith


Total Posts: 1
Joined: May 2020
 
Posted: 2020-05-25 07:15
2008 finance crisis,and bitcoin born;2020,covid19 crisis,maybe the baby will grown fast

nikol


Total Posts: 1172
Joined: Jun 2005
 
Posted: 2020-05-25 09:05
@keith
"2008 finance crisis,and bitcoin born;2020,covid19 crisis,maybe the baby will grown fast"

FED: "push, push, push"

gaj


Total Posts: 112
Joined: Apr 2018
 
Posted: 2020-06-03 05:11
SPX is nearly flat YTD, NDX is close to all time high. Do you guys think the stock market recovery is sustainable? I'm glad I stayed long equity. But how does it make sense that the stock market is close to where it was before the virus breakout?

EspressoLover


Total Posts: 445
Joined: Jan 2015
 
Posted: 2020-06-03 06:39
Here's my amateur thoughts on the bull case. The duration of major equity indices is 25+ years. Only a small fraction of most large-caps' NPV come from earnings over the next 4-8 quarters.

Let's assume a hard lockdown continues until the end of 2021, and corporate are zero until Q1 2022. How much should that impact valuations? How much should we discount stocks relative to their January highs? Based on DCF analysis, I think the answer is about 5-15%. Which implies that current prices are about fair. And that's not even taking into account higher P/E multiples that come with the reduction in the risk-free rate that almost assuredly accompanies that scenario.

The logical question is why doesn't this same reasoning apply to all recessionary bear markets. And I think the answer is that the typical recession is also accompanied by a clearing of mal-investment accumulated during the expansionary phase. The hangover not only comprises the loss of earnings during the recession, but the necessary trimming of fat from all the poor decisions that were made during the ebullience of the late-stage boom.

In 2007, the economy was allocating way too much capital to the finance sector, particularly mortgages, home building, raw materials, and unsustainable consumer spending powered by home equity. Citigroup, even today is still 90% below its 2007 highs. In 1999 the equity markets was in retrospect obviously throwing away capital into the giant money pit of the Internet bubble. In 1980, there were many business models and capital structures that were wiped out because they made sense at 8%, but not 16% yields.

This current recession seems very different. At one point the lock downs will be over, and there's not really any reason the economy and asset markets can't return to status quo antebellum. The recession wasn't caused by irrational exuberance, but an exogenous and temporary supply shock. Business models that worked in Q4 2019, should still work in Q1 2022. There might be some exceptions, where the culture kind of shifts permanently. Like air travel or commercial real. But by and large these sectors don't constitute a large share of large-cap US equities. I don't really see any 2020 equivalent to Citigroup or Cisco or Continental Illinois.

Good questions outrank easy answers. -Paul Samuelson

nikol


Total Posts: 1172
Joined: Jun 2005
 
Posted: 2020-06-03 07:18
@gaj

> Do you guys think the stock market recovery is sustainable?

It means that all printed money are channeled through stock markets... Ask yourself if it matches all going down indicators of real economy (international trade, jobless, less retail activity, necessity to shrink production to maintain prices)

AB12358


Total Posts: 67
Joined: Apr 2014
 
Posted: 2020-06-03 23:21
NPV with <=0 RFR is infinite, right? :D

svisstack


Total Posts: 355
Joined: Feb 2014
 
Posted: 2020-06-04 00:11
@EspressoLover: Did you think we can apply any models to the current situation?, especially very simple ones like dcf?

First Commander of the USS Enterprise
Previous Thread :: Next Thread 
Page 4 of 5Goto to page: 1, 2, 3, [4], 5 Prev Next