Youyang Gu

3.7K posts

Youyang Gu

Youyang Gu

@youyanggu

Data Scientist. Creator of https://t.co/ByO92M3ceQ and https://t.co/k9yZN1fORq. Presenter of unbiased takes. Realist. @MIT '15.

New York, NY Katılım Şubat 2009
616 Takip Edilen65.3K Takipçiler
Sabitlenmiş Tweet
Youyang Gu
Youyang Gu@youyanggu·
It's popular to use the US as a punching bag these days, but I can't think of many other countries where the work of an untrained 27-year-old immigrant can garner the attention & respect of researchers and public alike. America isn't perfect, but I'm grateful to call it my home.
English
121
740
8.9K
0
Youyang Gu
Youyang Gu@youyanggu·
Global excess deaths in 2020-2021 is almost 3x higher than reported #COVID19 deaths. Great work by @WHO & @UNDESA to get this out. It's probably the most comprehensive estimates to date. twitter.com/WHO/status/152…
World Health Organization (WHO)@WHO

The total number of deaths associated with #COVID19 worldwide from 2020-2021 may be closer to 14.9 million: New estimates by WHO & @UNDESA. That’s 9.5 million more deaths than reported bit.ly/38SJIpM #HealthData

English
15
18
86
0
Trisha Thadani
Trisha Thadani@TrishaThadani·
🚨TODAY a year-long investigation by @JoaquinPalomino and I publishes:We examined San Francisco's long practice of housing thousands of its most vulnerable homeless in run-down, century-old hotels called SROs In many cases, we found disastrous results sfchronicle.com/projects/2022/… 1/
English
25
173
369
0
Youyang Gu
Youyang Gu@youyanggu·
Happy Monday. I enjoyed my time on the Data Science Show with @DalianaLiu to talk data, Covid and lessons learned along the way. If you want to get into data science, I highly recommend giving Daliana a follow. youtube.com/watch?v=roRAhB…
YouTube video
YouTube
English
2
3
41
0
Youyang Gu
Youyang Gu@youyanggu·
@MaxCRoser that you have to start *somewhere*. So that leads to an interesting discussion about short-term vs long-term impact, which is probably the basis of longtermism. I've always been a more short-term > long-term impact guy but I like how you tie them all together as inter-connected!
English
0
0
1
0
Youyang Gu
Youyang Gu@youyanggu·
@MaxCRoser Ah thanks for the wonderful insights (as always). Will definitely share this with my friend and continue the discussion. I imagine someone in 1920 could probably be more "effective" in working in global/public health than say driving down solar costs. But I see your point that..
English
1
0
3
0
Youyang Gu retweetledi
Max Roser
Max Roser@MaxCRoser·
My new post is out: ourworldindata.org/longtermism It is about humanity's past and future. I don't know how to summarize this post in a thread. But I can share the two visuals I made for it. 👇
English
16
166
547
0
Phoebe Yao
Phoebe Yao@phoebeyao·
The early believers are your stepping stones: I founded my startup with a $1K grant (@1517fund) Dropped out of college with an $8K grant (@tylercowen) Built our team with a $100K grant (@thielfellowship) Today, we announced our $4.5M seed raise!
Pareto.AI@hellopareto

We’re excited to announce a $4.5M seed raise led by @MaCVentureCap. After 2 years of working with startups to supercharge their operations, we’re growing our team to develop the best virtual analyst for scaling businesses. Read more about our journey here: pareto.la/3JdW7lk

English
63
34
514
0
Youyang Gu
Youyang Gu@youyanggu·
@BillHanage @NateSilver538 Odd response from you. If you're only analyzing & Tweeting examples where interventions are seemingly successful (aka the survivors), and overlooking those that aren't, then that's the definition of survivorship bias. Not sure what "I do this for a living" have to do with it.
English
0
0
14
0
Nate Silver
Nate Silver@NateSilver538·
Australia has most of its population under a lockdown so strict as to resemble the terms of house arrest and yet cases are still rising there. I'm not sure why it's example of any sort of success that the US would want to emulate.
Bill Hanage @BillHanage.bsky.social@BillHanage

Delta is really transmissible, which is much worse than immune escape. On the other hand, there’s reasons for optimism from the likes of Australia and Vietnam that we can take action other than vaccination to stand on its way. Let me explain… 2/x

English
378
507
3.8K
0
Youyang Gu
Youyang Gu@youyanggu·
@DKThomp I've been thinking about this too... thanks for sharing!
English
3
0
18
0
Derek Thompson
Derek Thompson@DKThomp·
Great chart of how the perceived decline in vaccine effectiveness might be driven, largely or in part, by Simpson's Paradox* *ie: in which efficacy vs. hospitalization for specific age cohorts is actually MUCH HIGHER than efficacy for the total population
Derek Thompson tweet media
English
29
175
772
0
Youyang Gu
Youyang Gu@youyanggu·
@BillHanage @NateSilver538 Also the below Tweet is quite relevant.
Youyang Gu@youyanggu

@CT_Bergstrom When people use mere correlation to suggest that restrictions are effective, you see scientists nod their heads. But when people use correlation to suggest restrictions aren't effective, scientists immediately scream "wrong". I just don't think that's very scientific, IMHO.

English
2
0
8
0
Youyang Gu
Youyang Gu@youyanggu·
@BillHanage @NateSilver538 I specifically made sure to read the whole thread before making that reply, but thanks for asking! We both know that cherry-picking two countries & two states don’t do much to prove a claim. But alas it’s your Twitter so I respect your decision. Thanks for the “gotcha” though!
English
1
0
12
0
Youyang Gu
Youyang Gu@youyanggu·
@Ibeenhacked @EricTopol States and counties may be slower to report breakthrough cases, so that may skew the results, hence the ratio going down over time. Also, clusters could start in unvaccinated pockets and migrate to more broad regions.
English
1
0
1
0
Matt Jolley
Matt Jolley@Ibeenhacked·
@youyanggu @EricTopol The best denominatgor is per 100,000 vaccinated, or per 100,000 not vaccinated. Why would you expect ratio of case rates to change over time unless the denominators changed?
English
1
0
0
0
Eric Topol
Eric Topol@EricTopol·
These are the kind of data that need to be communicated to the public, as tracked by Oregon Yes, with Delta the breakthrough cases have increased but the benefit of vaccination has gone up markedly oregon.gov/oha/covid19/Do…
Eric Topol tweet media
English
75
869
2.5K
0
Youyang Gu
Youyang Gu@youyanggu·
@Ibeenhacked Hi, I was wondering if you could elaborate on this: "Maybe best not to use to guess vaccine efficacy"?
English
1
0
0
0
Matt Jolley
Matt Jolley@Ibeenhacked·
California breakthrough Covid cases trends. 15,700 cases week ending August 7th. 3 “weekly” reports of average daily cases per 100,000 vaccinated 3.5, 7, 8.2 “not vaccinated” 20, 33, 51 Maybe best not to use to guess vaccine efficacy
English
1
0
0
0
Matt Jolley
Matt Jolley@Ibeenhacked·
@youyanggu @EricTopol Thanks to for database link and vaccination rates. twitter.com/NuritBaytch/st… And to the people who suggested selected communities had higher vaccination rates than national rates.
Nurit Baytch @NuritBaytch

The data coming out of @IsraelMOH on the efficacy of the Pfizer vaccine against delta variant is alarming. the file can be downloaded here: data.gov.il/dataset/covid-… below is an image showing the % of Israelis fully vaccinated (column C) by age group @deeptabhattacha @GuptaR_lab

English
1
0
0
0
Youyang Gu
Youyang Gu@youyanggu·
@Ibeenhacked @EricTopol Wow that's quite interesting that the results are so different. Why do you think VE is much lower in Israel? And do you have the data for the 2444/492 breakdown? Thanks for doing the calculations.
English
2
0
1
0
Matt Jolley
Matt Jolley@Ibeenhacked·
@youyanggu @EricTopol That is quite a rabbit hole to go down, what possible efficacy with what possible vaccination rate could produce 83 percent breakthroughs compared to 17 percent of cases no shots. Assume a 90 percent rate. That would mean 492 cases from 10 percent of 1.15 million. About 46% VE.
English
2
0
0
0
Youyang Gu
Youyang Gu@youyanggu·
@jgharris7 @EricTopol Possible, tho vaccinated people are also less likely to get tested, so that would lower the effectiveness. But like I said, it's a rough estimate.
English
2
1
3
0
Youyang Gu
Youyang Gu@youyanggu·
@LeePDunham1 I think poverty rate was one of the variables, but it wasn't as correlated.
English
1
0
1
0
Lee P. Dunham
Lee P. Dunham@LeePDunham1·
@youyanggu Thanks! If you adjusted for that by using income over a regional CPI (or something), would adjusted income be better correlated? It seems to me that overall poverty would be more related to poor health outcomes than the Gini coefficient, so if that's not so it's very interesting.
English
1
0
0
0
Youyang Gu
Youyang Gu@youyanggu·
I believe income inequality is the single best predictor of total Covid deaths in the US. Not income, but income *inequality*. The R^2 is surprisingly high: 0.35. Thanks to @joe_sill, I was able to look at over 40 different variables, and expands on my analysis from last month.
Youyang Gu tweet media
English
56
172
533
0

Keşfet

@WHO @UNDESA @TrishaThadani @JoaquinPalomino @DalianaLiu @MaxCRoser @phoebeyao @1517fund