🚨In Guatemala,the world's main cocaine corridor,criminals deforest to smuggle cocaine
Using IV, I show cocaine seizures in Colombia lead to ⬆️ violence in Guatemala as criminals compete for more lucrative drug market
⬆️Emigration from Guatemala to Mexico
Fabolous book on
#Python
and
#Econometrics
It covers:
-basic python libraries,
-graphs,
-OLS,
-time series, forecasting,
-panel data,
-IV,
-simultaneous equations models,
-limited dependent variable models,
and various examples from
@jmwooldridge
book
⏰New paper: Chained Difference-in-Differences estimator for long-term treatment effects using unbalanced panel data. It handles multiple time periods, treatment timing variation, heterogeneity, and missing data patterns.
Cool paper estimates the per capita income of Aztec empire in 1492 at $690 (comparable to income of Euro-Mediterranean in 150ad) & Gini index of income inequality of 50.4, showing high inequality in the empire existed prior Spanish conquest
Replication:
Lost in the new DID literature? These help big time:
-DID lit survey by De Chaisemartin and d’Haultfoeuille
- What’s trending in difference-in-differences? by Roth et al.
Huber's book on causal analysis breaks down hard-to-understand topics in a way that is easy to grasp. IMO suitable for undergraduate & postgraduate students.
So far, I've read the chapters on instrumental variables and selection on observables, amazingly clear & with R examples
I have the crazy idea of creating a reading group on econometric papers. Focused on new IV and DID lit.
Anyone interested? Starting in May Fridays 9-10am UK time.
I will not teach the material, just discuss papers share tips on replication files.
#EconTwitter
#AcademicTwitter
Have you read this DID paper?
DID estimators for treatments continuously distributed at every time period,as is often the case of trade tariffs, or temperatures.
OMG, I won the
#AEA5k
race in the female microeconomist category with 21:02 😍😍😍🤩 19 months ago I was bed ridden with covid, as a sack of potatoes. It took me sooo long to recover from it. I'm thrilled to have survived this pandemic with you.
I took proof-based math courses during my undergraduate in Econ. Starting these courses during PhD might be a bit too hard for some. But never too late to start or refresh. Here a free course on how to do math proofs at MIT👇 Tom Leighton is outstanding 🌟
I don’t care if it’s real analysis in particular. But one has to be deeply incurious not to take a proof-based math course before starting an econ PhD.
Superb, an updated version of the diff-in-diff estimator for cases where the treatment is continuously distributed at every period. Consider instrumental variable cases as well.
Stata and R packages computing the estimators will be available soon, some can already be computed by
Happy to release new version of "Difference-in-Differences for Continuous Treatments and Instruments with Stayers", with Xavier,
@FelixPasquier
, and
@gvazquezbare
! Thread giving you the gist of the paper available under nicer skies than in Paris.
I knew that de Chaisemartin and D’Haultfœuille were writing a textbook, but had not realised that the draft version is already available, for free, and 111 pages long!
Looks fab 🌟 on Diff-in-Diff, multiple treatments, Bartik IV designs.
Available here:
If you got lost in the jungle of recent developments in
#causalanalysis
based on difference-in-differences, you should get this (free) book (work in progress) by
@CdeChaisemartin
and Xavier D'Haultfœuille: 👍😀
This is one of my top 5 favorite econometrics books. It has outstanding, clear explanations and plenty of examples in
#R
and now
#Python
It covers all modern quant tools, from OLS to advanced IV, causal machine learning, diff-in-diff, changes-in-changes, synthetic controls,RD etc
Hi
#EconTwitter
! 📊
Interested in exploring the intersection of
#econometrics
and
#machinelearning
for causal inference?
Don't miss the online version of
@CausalHuber
's book! It offers an insightful overview of quantitative methods for evaluating causal effects
Bonus:
🚨Several fully-funded PhD scholarships at QMUL 🥳
I welcome applications from qualified candidates interested in working in econ development, inequality or organised crime.
Send me a DM if interested
Requirements:
Queries: sbmphd-admissions
@qmul
.ac.uk
Paper found 63% of female academics receive abusive comments in student evaluations vs 52% of men.
The abuse impacts academics' well-being and career progression.
Some unis are removing comments from the evaluations. Does that solve the problem? 🤔
#AcademicChatter
I start the week by re-reading this amazing diff-in-diff paper.
@assamidanov
&
@edouardmensah
will present it in
#METRICS
this Friday.
More details soon
Stata and R replication files
@Daverober
@danielaivetteas
David, siento que el editor te haya enviado semejante dictamen.
Haznos a todos un favor: Envía una queja al comité editorial.
No podemos tener a evaluadores enviando dictámenes tan insultantes.
Morning!
Today I'll be sharing 15
#productivity
tips & what I do during my day from 6am until 11pm.
I'm not teaching today, have no meetings. So this is how my full research days look like.
Tip 1: Have a schedule hour by hour and try to stick to it.
#AcademicChatter
@TansuYegen
In my swim class this week, we were talking about this windmill style.
In contrast to some of my swim classmates, I much prefer Michael Phelps' high elbow recovery to the windmill. Ledecky's arrow style is also incredibly smooth and efficient.
Over 500 local politicians have been killed in recent years in Mexico. In this working paper, we analyze the role of organized crime and its impact on voting behavior.
I'll present these findings on Friday, at the Conflict and Contentious Politics Workshop at Essex University
🇲🇽 30 candidates have been murdered as Mexico prepares for general elections on Sunday.
Hundreds have dropped out or asked for protection as criminal groups target them to push their political interests.
@Theholisticpsyc
Someone stole property from me. I hired a lawyer to recover it. The thief told me that I hold grudges, that is bad for my health, and that I should not feel annoyed at all. 😂😂The level of gaslight.
@_danigilbert
Same here. It speaks volumes about the sender's biases. My strategy nowadays is to ignore these emails, particularly if they were expecting a reply from me. It works wonders. Suddenly my titles re-appear in the correspondence.
Happy first day of
#AcWriMo2021
everyone!
Goal this month is to read & write a lot with two fabulous e-writing groups and wonderful
@raulpacheco
#AcWriMo
✍️✍️
Have you received the email from USS?
"Pay your pension some attention"
😂 I've indeed been paying attention to the non-existent USS pension deficit vs our pension cut.
Last week, I attended the
#USS
webinar about the changes in the pension scheme. The webinar was really really bad. Confusing, misleading, and a waste of our time.
This webinar 👇about the pension changes by Sam Marsh from UCU is clear, to the point, and definitely worth watching
I gave a half-hour update on
#USS
to
@oubucu
yesterday, trying to cram as much in as possible. If you need to get up to speed on things ahead of the upcoming strikes and reballot, then you can watch it here:
Les compartimos un hilo de nuestro panel hoy en
#OC24
sobre violencia de género:
Estudios sobre violencia en México utilizan generalmente indicadores de homicidios. Este proxy no contempla los varios tipos de violencia y crímenes a los que están expuestos mujeres y hombres. 1/10
@johnwhitehead81
Worse, last year I got a paper rejected by reviewer2 because argued my logit model with a pseudo-R2 of 0.15 was leaving 0.85 unexplained and such model would be pretty useless for policy advice. I gave up defending the paper from such reviewer and resubmitted elsewhere.
If anyone is interested in forming/joining a virtual reading group focused on
#violence
in
#LatinAmerica
send me a DM or email r.gutierrez
@qmul
.ac.uk
I've been proposing this to
@raulpacheco
. Difficult to find time, so might just start something small
"No toda la violencia que ejerce el crimen organizado es letal. Eso es lo que se ve en los medios de comunicación.
Se ejerce una violencia de intimidación y amenazas de manera regular..."
@ahope71
Foro: elecciones estatales y crimen organizado.
Kenya has experienced low-intensity but recurrent electoral violence since 90s.Using DID I show areas affected by such violence have
⬆️30% food inflation for a trimester
⬇️in night light for a year
Using matching,I find children exposed to violence have stunted heights as adults
I start reading another amazing diff-in-diff paper by Roth and Sant'Anna.
@FrancescaFolia1
&
@RaisaTSara
will present it in
#METRICS
on Friday, 25 November 🥳
Stata and R replication files
Free books that provide a gentle introduction to
#R
,
#Python
and
#Julia
, all based on Wooldridge's excellent "Introductory Econometrics"
Link to download:
All of them are available on Amazon. I have the first two but had no idea about the Julia version
@JoshGoodman_BU
@Prof_Nishith_P
Also, last year I ran a reading group focusing on the latest DID methods. We studied some of the publicly available replication files for each method. You can find the material here
For the past 15 months, I've been working on a paper on an irregular basis. Every time I retake it, I tell myself that I'll finish it this week. Tomorrow could be the day.
#GetyourManuscriptOut
#WishMeLuck
🤞
🚨WP with
@mikk_c
Network analysis finds municipalities that experience a unit ⬆️in homicide rate leads
⬆️41 people emigrate domestically
⬆️28immigrate domestically
⬇️2Mexicans return from US
⬇️1 move to US
⬇️20kdaily🚗trips w rest of country🧵
htps://arxiv.org/abs/2301.12743
The 2023 ESRC Postdoctoral Fellowships competition is open 📢
7 posts at King’s College London, QMUL, or Imperial.
Deadline 23 March, details:
If you wish to work on political economy, conflict, or inequality at QMUL, drop me a line, happy to discuss.
🚨
@nayiturbel
@yanahilop
@dani_cristob
and I will present in the:
Panel on gender violence and organised crime
at the 24-hour Conference on Global Organized Crime
#OC24
! click below to sign up and view the schedule
#OC24
🧵preview about panel
I'll be releasing details about the metrics reading group in a few weeks.
I thought that there would be about 4 people interested 🙊 but it turned out that there is much higher demand.
Once the registration website is set up, with all the reading material, I'll get back to you
Today in
#PEACELA
we read 👇
The article aims at explaining why drug cartels have assassinated hundreds of local elected officials, state government officials and party candidates in Mexico.
A 🧵of key findings and our discussion
Happy 1st September 🥳
This month I hope to finish the two papers I've been working on for 1 1/2 years.
Big data projects on political violence and femicides in Mexico.
Dengue is fastest growing mosquito-transmitted disease putting 40% global pop at risk due difficulty in identifying timing/areas most at risk.
New paper
@CCarretoR
@sheis_andromeda
& I predict daily risk of transmission of
#dengue
#chik
#Zika
in Mex 🧵
One of my PhD students is analysing harassment and bullying on Twitter.
We were talking about it last week, where she could get data, design an RCT psychological intervention to reduce harm and anxiety. Neither of us anticipated what would be going on in econtwitter. Jeez
Gender violence is on the rise in many countries as evidenced by the alarming number of femicides (hate killings of girls & women because of their gender).
To reduce such killings, 18 countries in Latin America and the Caribbean have introduced femicide laws.
Do they work?🤔
Last December I submitted 4 grant proposals to 🎅
I got now the response from all:
1 ☑️ 🥳
2 ❌ ok, knew had a very small chance
3 ☑️ 🥳
4 ❌ 😞nope, but will try again
The silver lining 50% success rate, must be getting less bad at asking £££
@CavaliereGiu
@causalinf
@jmwooldridge
@lucafanelli5
Thanks for sharing. This website 👉 has info on the STATA code to add tF critical values and standard errors to ivreg2:
ssc install ranktest
ssc install ivreg2
net install tf, force from()
help tf
Reviewer 2, in nth round of revisions, asks:
Why I chose NOT to take the log of covariate X, which clearly takes negative, positive and zero values 🫠
Should I elaborate as a footnote what happens when attempting to take the log of a var X<=0?
Guess what just arrived?
I'm particularly interested in learning more about doubly robust methods, causal machine learning, and synthetic controls. The diff-in-diff chapter is updated to the latest published DiD methods literature. Fabulous 🥳
Fantastic new report
#genderimbalance
. Amongst academic economists in the UK, women comprise:
33% of lecturers
27% of senior lecturers/readers
15% of professors
No Black female professors of econ were employed between 2012-2018/9
The ups and downs of research. My october
2 articles submitted 🥳
1 rejected after 1 year under review. WT! 🤧
1 finally accepted 🥵
1 new project went to bin too much freeriding 🤐
1 article just invited for R&R 🤗
OMG, paper accepted at last!
We worked so hard on that paper, particularly
@CCarretoR
and another awesome young scholar Tania Rodríguez. I'm so happy.
We'll share the paper about dengue and climate in Mexico soon.
I've spent two days coding to produce this graph.
The policy that I've been analysing with
@nayiturbel
has no impact on reducing homicides of women in Mexico.
I was expecting *some* impact, but we get nothing.
📢Several PhD scholarships available at QMUL.
Entry req 👉
Scholarship details:
Happy to answer queries from students wishing to work on econ dev or conflict.
#Scholarships
#PhD
I look forward to presenting this paper co-written with amazing Michele Coscia
@mikk_c
We use network analysis and IV to quantify the impact of violence on migration using data from the last 4 censuses 2005-2020, and the change in vehicle trips 🚗🚗 in Mexico.
My first run in 9months post-covid, and first entry to
#aea5k
. 19k cycling 🚴(45min)+5k running wearing 😷 (30 min) in gorgeous yet freezing cold London. Thank you
@causalinf
for organizing this. Happy New Year!
This Wed 18 October, we present insights on femicides using new daily georeferenced data in Mexico.
Also, have new unilateral divorce laws shortened the time it takes to get divorced in Mexico?
And, what is the impact of femicide laws?
Book your place:
I think I've finally found a 🤩 python🐍 course online. If I go beyond the 5th project I'll share details.
My previous attempts: before covid, I took 2 face-to-face courses, v useless. Then 1 way too advanced on ML. Then 2 e-courses, which I quit after spending weeks on print()
You can find more info about the book at this link
In that link, the book can be read online by clicking on the option "here as a HTML online book"
The exercises and scripts are available on the same website, just click on Downloads
Happy to support eligible candidates who would like to have their fellowship based at QMUL, London, working on economic development, gender-based violence, conflict resolution, and inequality. Drop me an email: r.gutierrez
@qmul
.ac.uk.
We’ve opened the call for
@BritishAcademy_
Visiting Fellowships 2025!
As part of our commitment to international engagement we will give up to £40k for academics to spend 3-6 months to be based at a UK higher education or other research institution
This new website to help writing is amazing 😍
I entered the phrase "Academic writing is hard"
Then the website returned a full paragraph based on my phrase.
#AcademicTwitter
#AcademicChatter
#PhDChat
Today we launched YouWrite — an AI writing assistant built into .
For me, this is a special moment that ties my past research to the vision.
YouWrite example below👇with thoughts on search, research, writing, and meaning. 🧵
#METRICS
reading group discusses this Friday 🥳
Sun & Abraham
#DiD
paper 👇
@rama_mosh
will present the theoretical insights &
@AidaGL_econ
will talk through replication code
We'll cover 🧵
@Daverober
@danielaivetteas
Mucha suerte David. En 20+ años de carrera me he quejado una sola vez de un dictamen. Mi carta fue: Sugiero al comite leer el dictamen, el tono es iriente e innecesario.
El editor me envió una disculpa extensa por no haber leído el dictamen antes y prometió bloquear al evaluador
Amazing six and a half hours of diff-in-diff workshop with
#codechella
and
@causalinf
. Part 2 next week. Scott will cover other 5 new papers on diff-in-diff, including fuzzy diff and more R, stata code
Delighted to host all these wonderful researchers and looking forward to
#PuzzleofPeace
on Monday-Tuesday - several days of amazing research on conflict, violence and peace
The readings of
#PEACELA
👇
All books & readings on gender violence are outstanding. Learned tons about ethnography, environmental violence. Readings on political violence & criminal organizations paint a depressing picture. So much corruption and despair
This November in
#AcWriMo
, I said that I was going to focus on finishing one and one project only. The article I've been working non-stop for two years.
How is that going?
I'm submitting it later today!!!
#WritingCommunity
A partir de hoy entran en vigor las reformas al Código Penal Federal y a la Ley General de Salud que prohiben las mal llamadas “terapias de conversión” o
#ECOSIG
.
Ayer fueron publicadas por el Ejecutivo en el
@DOF_SEGOB
.
#NadaQueCurar
🏳️🌈🏳️⚧️
Tomorrow, 18 Oct, 19:00pm UK, we present a new femicide dataset for Mexico.
The data allow us to conduct analysis at a very small-area level. The spatial concentration of these killings is horrific in some areas.
Come and join the presentation:
This Wed 18 October, we present insights on femicides using new daily georeferenced data in Mexico.
Also, have new unilateral divorce laws shortened the time it takes to get divorced in Mexico?
And, what is the impact of femicide laws?
Book your place:
🚨
#Metrics
meets next 4th November to discuss another great diff-in-diff paper
@hediehtj
will present the theoretical insights &
@Saad7_aftab
the replication files
We look forward to it 🥳
Our reading list with replication files
I sent two v small, tiny, grant proposals last month.
Both proposals have been granted 🥰
One of the sponsors had rejected my proposals a few years ago. I learned a lot from the feedback even though it saddened me back then.
#grants
#academicWriting
#Motivation
🚨New Publication. Experiment challenges conventional views on ethnic voting 👇
Study in Kenya shows voters have incorrect perceptions of how their MP spent local public funds. Voters' ethnicity and partisanship don't affect these misperceptions, then 🧵
@raulpacheco
Yes, my term is 12-weeks long, and week 7 is reading week. In that week there are no assignments, no lectures, no seminars, no office hours. It is a superb break for everyone.
@ben_golub
My favourite on this is Tom Leighton MIT course. Fantastic mathematician and teacher of how to do proofs. Video lectures, lecture notes, excercises and exams with answers available for free here
@jeffguhin
I survived long-covid and recovered after 9 months of being a couch potato. I can now exercise continuously for 5 hours on weekends (cycle, run, swim, squash, weights), and enjoy it too.
We had the first
#METRICS
meeting today!
Topic 1: Bacon decomposition:
To understand the paper, we discussed:
@agoodmanbacon
amazing explanation he gave a while ago
and example by
@AsjadNaqvi
Michoacán exports berries & half of world's avocados, earning $7billion a year. Criminals extorted the producers, who fought using armed defence groups
Moncada’s book shows how firms in LATAM resist extortion & limitations of defence groups
My book review
Hi, for those of you who signed up for the
#Metrics
reading group 🥳
Remember that today is the deadline to confirm your participation in the group.
If any spaces become available, I'll contact interested people currently on our long waiting list 🙌