Tulio
@tuliosousa93
Followers
200
Following
4K
Media
0
Statuses
130
Econ PhD candidate @DukeEcon, incoming AP @KelleySchool. Broad research interest in Industrial Organization and Microeconometrics. Digital Ads and Data Privacy.
Durham, NC
Joined October 2022
I often work with young researchers who don't convert R&Rs to publications. I definitely don’t do everything well! But I put a lot of effort into this and have a 24/24 conversion rate. So I am passing along six tips based upon my experience as an author and editor. #Econtwitter
15
279
1K
1. Who pays for tariffs? It’s a simple question that’s surprisingly tough to answer. We often see only a rough aggregate measure of prices at the border or the retail price, missing many steps in the chain.
6
43
148
Today, @CMAgovUK decided to release Google from its Privacy Sandbox-related commitments. Read more:
1
2
5
Finally, someone has solved a real problem with AI! No more having to take a paper in the format for a journal that rejected you, and reformat it for a new journal. Well done!! https://t.co/J5HekY1Pc0
47
442
3K
80% of all digital ad sales go through “closed loop” auctions where one big player like Google/Meta/Amazon runs the whole process. The industry wants sellers to voluntarily disclose how they determine prices and winners in auctions that manage around $750B each year
For years advertisers haven’t known how the biggest digital auctions actually work. Industry players want to finally change that.
1
7
18
The casual browsing activity across the open web is diminishing at an exponential rate. As a result, logged-in user data and retail media IDs are increasingly becoming the core "data spine" for advertising and targeting. Even if third-party cookies were killed it wouldn't matter
0
2
19
So you want to run an experiment....now what? We are moving through the optimal experimental design parts of the textbook in today's Chicago School of Experimental Economics lectures. From the principles of optimal design emerge several handy rules of thumb, some of which
ideas.repec.org
Experimental economics represents a strong growth industry. In the past several decades the method has expanded beyond intellectual curiosity, now meriting consideration alongside the other more tradi
1
11
72
If you're an economist and haven't heard about the double descent phenomenon, you might be overlooking one of the most interesting developments in computer science and statistics today. Personally, I haven't come across anything as fascinating since I first learned about Markov
38
199
1K
Common ownership isn't just an academic curiosity anymore. Isabel Tecu and I have a new guide for antitrust practitioners reviewing the theory, evidence, mechanisms, and policy debates around investors holding large stakes in rival firms. Comments welcome!
4
24
120
.@Google real-time bidding privacy settlement will require company to give users new controls over online tracking/ad targeting.
2
5
13
A key factor in dominating the buy side of programmatic advertising is the ability to process high volumes of QPS. However, not all QPS is treated as equal in value to the DSP. Even the largest DSPs discard QPS they deem unlikely to result in transactions, prioritizing factors
2
1
16
I'm putting together a talk on a new applied micro paper, and I'm reminded that (I believe) it's very hard to outperform the structure proposed in this piece by Jesse Shapiro. https://t.co/LY0Tj7aWIZ
3
83
567
Normally, it's: 1) write a paper & submit 3) get reviews (~3 months) 4) revise paper & resubmit 5) wait for response (~3 months) ...what if we could simulate this process in minutes? Could we fix issues? Anticipate misconceptions? Get ideas for new analyses/experiments? 1/
10
35
195
We planned this podcast "debate" about Transaction IDs a couple of weeks ago, and the morning we're recording it the ground shifts and everything gets thrown into chaos. Fun stuff. cc: @ckane @mikeosullivan
New pod: Chris Kane and Mike O’Sullivan debate publishers' adoption of Transaction IDs and whether they’ll even be able to with the latest changes to prebid. Link below: 👇
1
3
15
All of adtech is fighting over identity, both in the form of actually bidding on auctions and in the form of QPS. It all comes down to user IDs, which enable platforms to apply their learning and use that learning to bid on auctions targeting their desired customers. QPS without
6
2
30
Sim, secretário, o Brasil é um país extremamente fechado—em praticamente qualquer comparação. (Gráfico abaixo do livro “Política comercial no Brasil”, de Ornelas, Pessoa e Ferraz)
1/- Sugestão de reflexão sobre a tese do livro O Brasil realmente é uma economia fechada? Ou a tese não passa de senso comum? A tarifa efetiva média do Brasil é de 4,9%. Exemplo: a relação comercial com os EUA,a tarifa efetiva média é de 2,7%. Segue. https://t.co/F7bb2aptq5
5
37
303
I recently emphasized this point, but it bears repeating: display ads are not disappearing, though the way we think about and purchase them is poised for a significant shift. AI-powered search is not a future prospect; it is here, fundamentally altering how users engage with the
2
1
13
Deforestation in the tropics will be a key issue at the upcoming @UN climate conference @Cop30noBrasil. In our latest policy brief, SIEPR Senior Fellow @BardHarstad outlines how a proposed new global fund designed to help finance conservation could be strengthened.
1
2
4
VoxDev has a great summary of our new paper. The lesson is also relevant for the design of domestic offset market for China’s national ETS.
🆕 Paying to pollute: How carbon offsets actually raised emissions in China Today on VoxDev, Qiaoyi Chen, Nicholas Ryan (@YaleEconomics) & @DanielYiXu (@DukeEcon) outline how a carbon offset programme raised emissions in China:
0
9
35
AI is the biggest buzzword in adtech right now. but what’s really happening with AI in major DSPs today? Let’s break down key tasks and see where AI stands. 👇
3
6
33