Last synced from Twitter 14m ago
Bbbreaking News: Discovering Amateur News Videos by Monitoring Journalists on Twitter
In August, I tweeted an idea for a breaking news site that monitored reporters asking for broadcast rights on Twitter. @probablycorey built it, and it's great.
Interaction data from the Copenhagen Networks Study
Measurement(s) Call • Text Message • facebook friendship • Proximity Technology Type(s) mobile phone • social network analysis • bluetooth Factor Type(s) timestamp • sex Sample Characteristic - Organism Homo sapiens Sample...
The gene-based hack that is revolutionizing epidemiology
Mendelian randomization offers a simple way to distinguish causation from correlation. But are scientists overusing it?
syllabus Marketing Analytics 2020 draft
15.819 Marketing Analytics MIT Sloan School of Management Spring 2019 TTh 1:00–2:30 (A) and 2:30–4:00 (B) Instructor: Dean Eckles <[email protected]> Teaching assistants: Madhav Kumar <[email protected]> Lead TA Leon Yao <[email protected]> Please put “15.819” in the...
2
Dean Eckles
Draft syllabus for my spring @MITSloan Marketing Analytics course docs.google.com/document/d/1T_…
One thing I'm trying more of this year: contests! Students will compete in both predictive modeling and also finding optimal targeting policies. There will be prizes: pic.twitter.com/Mli4EXBL8i
39h
Deep Generalized Method of Moments for Instrumental Variable Analysis
Instrumental variable analysis is a powerful tool for estimating causal
effects when randomization or full control of confounders is not possible. The
application of standard methods such as 2SLS, GMM, and more recent variants are
significantly...
Sensitivity Analysis via the Proportion of Unmeasured Confounding
In observational studies, identification of ATEs is generally achieved by
assuming "no unmeasured confounding," possibly after conditioning on enough
covariates. Because this assumption is both strong and untestable, a
sensitivity analysis should be...
Before We Kill The Cookie, Let’s Ask Why
"Data-Driven Thinking" is written by members of the media community and contains fresh ideas on the digital revolution in media. Today’s column is written by Kurt Donnell, president at Freestar. Right or wrong, the general public associates the term...
China Uses DNA to Map Faces, With Help From the West
Beijing’s pursuit of control over a Muslim ethnic group pushes the rules of science and raises questions about consent.
Google and Twitter Changed Their Rules on Political Ads. Why Won't Facebook?
'They have tried to stand for the principle of free speech,' says one expert, 'but the political and social reality of the situation is that it comes off tone deaf.'
California Homeowners Have 20 Uninhabited Bedrooms for Every Homeless Person
The search for housing justice shouldn't ignore the empty luxury spaces we subsidize most.
The origin of the Wilkinson-style notation such as (1|id) for random effects in mixed models formulae in R
Model formulae in R such as
y ~ x + a*b + c:d
are based on the so called Wilkinson notation: Wilkinson and Rogers 1973, Symbolic Description of Factorial Models for Analysis of Variance.
This pa...
Pete for America
It's time for a new generation of American leadership.
Dean Eckles
Voters don't type peteforamerica.com, they type "butt" and use autocomplete / wonder about their browser history.
21d
Andrew Yang for President - Humanity First
Humanity First
Dean Eckles
Candidates keep saying URLs, but people don't do that...
They don't type Yang2020.com, they type "yang".
They don't type amyklobuchar.com, they type "klob" and use autocomplete.
#DemDebate
21d
Amy’s Making A Big Announcement
Attend her big event in Minneapolis on February 10!
Dean Eckles
Candidates keep saying URLs, but people don't do that...
They don't type Yang2020.com, they type "yang".
They don't type amyklobuchar.com, they type "klob" and use autocomplete.
#DemDebate
21d
An update on our political ads policy
An update on some changes we’re making to how we handle political ads on our platforms globally.
[80] Interaction Effects Need Interaction Controls
In a recent referee report I argued something I have argued in several reports before: if the effect of interest in a regression is an interaction, the control variables addressing possible confounds should be interactions as well. In this post I...
1
Dean Eckles
Another way of saying this: Don't rely on restrictive parametric models of the outcome when doing observational causal inference. That's a big part of the appeal of, e.g., matching.
datacolada.org/80 @uri_sohn
22d
There Are Two Rhodes Scholars Running For President. Only One Gets Mentioned.
Media outlets mention that Pete Buttigieg was a Rhodes scholar far more often than they mention that Cory Booker was one too.