Dean Eckles

@deaneckles

networks, contagion, causality. professor

Cambridge, MA
deaneckles.com
Vrijeme pridruživanja: veljača 2007.
785 fotografija ili videozapisa Fotografije i videozapisi

Tweetovi

Blokirali ste korisnika/cu @deaneckles

Jeste li sigurni da želite vidjeti te tweetove? Time nećete deblokirati korisnika/cu @deaneckles

  1. Prikvačeni tweet
    prije 19 sati

    Which products are purchased together can reveal a lot about their relationship. But so can which products are viewed together, but not co-purchased. Our ( & ) new working paper leverages both types of baskets...

    34 proslijeđena tweeta 140 korisnika označava da im se sviđa
    Prikaži ovu nit
    Prikaži ovu nit
    Poništi
  2. proslijedio/la je Tweet
    prije 19 sati

    "last month it announced that ICE could go ahead and start destroying records from Mr. Trump’s first year, including detainees’ complaints about civil rights violations and shoddy medical care."

    The National Archives is letting millions of documents, including many related to immigrants’ rights, be destroyed or deleted.
    2 proslijeđena tweeta 2 korisnika označavaju da im se sviđa
    Poništi
  3. proslijedio/la je Tweet
    prije 14 sati

    Interesting findings from baskets of co-purchased & viewed products. — basket specifc embeddings help create bundles of complimentary/ substitutable products — ways of strategically expanding user’s taste: build cross-category bundles to get user to shop in new categories

    Which products are purchased together can reveal a lot about their relationship. But so can which products are viewed together, but not co-purchased. Our ( & ) new working paper leverages both types of baskets...
    Prikaži ovu nit
    3 proslijeđena tweeta 8 korisnika označava da im se sviđa
    Poništi
  4. proslijedio/la je Tweet
    prije 15 sati
    Odgovor korisniku/ci

    Indeed — I use materials from this impressive meta-analysis of hundreds of ad experiments when teaching marketing analytics

    1 reply 2 proslijeđena tweeta 9 korisnika označava da im se sviđa
    Poništi
  5. proslijedio/la je Tweet
    prije 14 sati

    One group where the crime rate really is higher than in comparable populations

    When you hear Trump talking about the crime rate among immigrants, remember that his national security advisor, campaign manager, deputy campaign manager, longtime political advisor and personal attorney are all criminals.
    1 reply 9 proslijeđenih tweetova 62 korisnika označavaju da im se sviđa
    Poništi
  6. prije 15 sati

    In the past, I've assigned this very useful paper on ad measurement: But I think it is getting a bit out-of-date, including some of its (well-founded) pessimism. What's the best updated overview of learning about ad effects?

    1 reply 1 proslijeđeni tweet 18 korisnika označava da im se sviđa
    Poništi
  7. proslijedio/la je Tweet
    prije 15 sati

    Looking forward to seeing everyone at my talk tomorrow (at Harvard, at noon in CGIS 354): "Statistically Valid Inferences from Privacy Protected Data" Some slides:

    9 proslijeđenih tweetova 36 korisnika označava da im se sviđa
    Poništi
  8. proslijedio/la je Tweet
    prije 15 sati

    Very interesting work from very talented people. Especially if you're in retail make sure to take a look.

    Which products are purchased together can reveal a lot about their relationship. But so can which products are viewed together, but not co-purchased. Our ( & ) new working paper leverages both types of baskets...
    Prikaži ovu nit
    3 proslijeđena tweeta 15 korisnika označava da im se sviđa
    Poništi
  9. proslijedio/la je Tweet
    prije 16 sati

    Proud of this work, & & his brilliant persistence! It's interesting to see the lit on "multi product pricing" both from a theoretical economic perspective & a data driven preference revelation perspective. Our larger goal is to link the literatures so we learn.

    Which products are purchased together can reveal a lot about their relationship. But so can which products are viewed together, but not co-purchased. Our ( & ) new working paper leverages both types of baskets...
    Prikaži ovu nit
    5 proslijeđenih tweetova 17 korisnika označava da im se sviđa
    Poništi
  10. prije 18 sati

    If you want to see what products are closest to Neutrogena Oil-Free Acne Wash Redness Soothing Cream Facial Cleanser, 6 Fl. Oz, in the purchase and view/search spaces, you will have to look at the paper:

    1 reply 1 proslijeđeni tweet 11 korisnika označava da im se sviđa
    Prikaži ovu nit
    Prikaži ovu nit
    Poništi
  11. prije 18 sati

    This paper reflects a lot of hard work by creating these dual embeddings and designing, running, and analyzing a field experiment with so many distinct "treatments".

    1 reply 1 proslijeđeni tweet 10 korisnika označava da im se sviđa
    Prikaži ovu nit
    Prikaži ovu nit
    Poništi
  12. prije 18 sati

    This includes some interesting "cross-category" bundles, which can be of strategic importance to a firm looking to expand the categories a customer shops in.

    1 proslijeđeni tweet 6 korisnika označava da im se sviđa
    Prikaži ovu nit
    Prikaži ovu nit
    Poništi
  13. prije 18 sati

    We use products' closeness as a heuristic to construct thousands of discounted bundles of products and try these out in a field experiment. Both scores positively predict responses to the bundles, especially the purchase-based complementarity score.

    1 reply 1 proslijeđeni tweet 9 korisnika označava da im se sviđa
    Prikaži ovu nit
    Prikaži ovu nit
    Poništi
  14. prije 18 sati

    Embeddings using either can capture many fine-grained categories. Here you can see existing categories in a 2D projection. Close products are predicted to be purchased together.

    1 reply 1 proslijeđeni tweet 9 korisnika označava da im se sviđa
    Prikaži ovu nit
    Prikaži ovu nit
    Poništi
  15. proslijedio/la je Tweet
    prije 21 sat

    Friends don't let friends draw almost any conclusions from observational nutrition studies ( bait)

    Today we have YET ANOTHER large cohort study showing an association between meat and death. It comes to basically the same conclusions, and has the same shortcomings, as other large cohort studies. When will we stop doing these?
    6 replies 12 proslijeđenih tweetova 39 korisnika označava da im se sviđa
    Poništi
  16. prije 19 sati

    For a rare binary variable, it is easy for measurement error to introduce bias orders of magnitude larger than the true %. ht

    There is no way 10 million Americans are nudists.
    5 proslijeđenih tweetova 25 korisnika označava da im se sviđa
    Poništi
  17. proslijedio/la je Tweet
    4. velj

    "robust" (Eiker-Huber-White) style standard errors are substantially biased downwards in many cases - especially those with a few "high leverage" data points, which Young 2019 finds in much of applied work. See Chesher and Jewitt 1987 for the proof:

    7 replies 21 proslijeđeni tweet 103 korisnika označavaju da im se sviđa
    Prikaži ovu nit
    Prikaži ovu nit
    Poništi
  18. 4. velj

    Thread on information cascades from a philosophical perspective

    Third, we read this paper by Hung and Plott which presents an experiment on information cascades. First, they confirm that humans, when placed in the same setting, behave largely like the mathematical model.
    Prikaži ovu nit
    4 proslijeđena tweeta 10 korisnika označava da im se sviđa
    Poništi
  19. proslijedio/la je Tweet
    3. velj

    What happens when the global economy tries to quarantine China? Tremendous deep dive by : In 2003, when SARS hit, China was a relatively small part of the world's trade. Today it's responsible for almost 1/5 of global GDP.

    24 proslijeđena tweeta 31 korisnik označava da mu se sviđa
    Prikaži ovu nit
    Prikaži ovu nit
    Poništi
  20. proslijedio/la je Tweet
    4. velj

    A coyote and a badger use a culvert as a wildlife crossing to pass under a busy California highway together. Coyotes and badgers are known to hunt together. 🎥Peninsula Open Space Trust

    5.169 replies 171.571 proslijeđeni tweet 723.951 korisnik označava da mu se sviđa
    Poništi

@deaneckles još nije objavio nijedan Tweet.

Čini se da učitavanje traje već neko vrijeme.

Twitter je možda preopterećen ili ima kratkotrajnih poteškoća u radu. Pokušajte ponovno ili potražite dodatne informacije u odjeljku Status Twittera.

    Možda bi vam se svidjelo i ovo:

    ·