Sai Chintala's Avatar

Sai Chintala

@saichandc

Assistant Professor of Marketing, LKCSB, SMU https://saichandchintala.github.io/

129
Followers
542
Following
8
Posts
05.12.2023
Joined
Posts Following

Latest posts by Sai Chintala @saichandc

Bayesian Data Analysis course - Aalto 2025 – Bayesian Data Analysis course

All the material for my Bayesian Data Analysis course is available online, including the lectures, which we re-recorded this fall (some of them by @aloctavodia.bsky.social and Noa Kallioinen while I was on vacation). The video links are listed in the schedule at avehtari.github.io/BDA_course_A...

11.12.2025 14:20 πŸ‘ 77 πŸ” 30 πŸ’¬ 0 πŸ“Œ 0
Preview
How Google Maps quietly allocates survival across London’s restaurants - and how I built a dashboard to see through it I wanted a dinner recommendation and got a research agenda instead. Using 13000+ restaurants, I rebuild its ratings with machine learning and map how algorithmic visibility actually distributes power.

I wanted dinner recommendations so I scraped 13,000+ London restaurants and accidentally discovered Google Maps is running a shadow economy. Anyway here's a dashboard and a political economy thesis: open.substack.com/pub/laurenle...

09.12.2025 07:52 πŸ‘ 850 πŸ” 351 πŸ’¬ 41 πŸ“Œ 93
Financial event studies, ubiquitous in finance research, typically use linear factor models with known factors to estimate abnormal returns and identify causal effects of information events. This paper demonstrates that when factor models are misspecified an almost certain reality traditional event study estimators produce inconsistent estimates of treatment effects. The bias is particularly severe during volatile periods, over long horizons, and when event timing correlates with market conditions. We derive precise conditions for identification and expressions for asymptotic bias. As an alternative, we propose synthetic control methods that construct replicating portfolios from control secu rities without imposing specific factor structures. Revisiting four empirical applications, we show that some established findings may reflect model misspecification rather than true treatment effects. While traditional methods remain reliable for short-horizon studies with random event timing, our results suggest caution when interpreting long-horizon or volatile-period event studies and highlight the importance of quasi experimental designs when available.

Financial event studies, ubiquitous in finance research, typically use linear factor models with known factors to estimate abnormal returns and identify causal effects of information events. This paper demonstrates that when factor models are misspecified an almost certain reality traditional event study estimators produce inconsistent estimates of treatment effects. The bias is particularly severe during volatile periods, over long horizons, and when event timing correlates with market conditions. We derive precise conditions for identification and expressions for asymptotic bias. As an alternative, we propose synthetic control methods that construct replicating portfolios from control secu rities without imposing specific factor structures. Revisiting four empirical applications, we show that some established findings may reflect model misspecification rather than true treatment effects. While traditional methods remain reliable for short-horizon studies with random event timing, our results suggest caution when interpreting long-horizon or volatile-period event studies and highlight the importance of quasi experimental designs when available.

Post image

Wrote a new paper on the econometrics of financial event studies, would value feedback! It's very new.

With my amazing grad student Tianshu Lyu www.tianshulyu.com, who is on the market. You should hire him!

paulgp.com/papers/finan...

19.11.2025 04:43 πŸ‘ 46 πŸ” 12 πŸ’¬ 1 πŸ“Œ 1
Preview
Python package managers: uv vs pixi? When I talk to people about Python package management in 2025 I see the following tools in active use; uv, pixi, pip, conda, mamba, micromamba and poetry. There may be others, but I don’t hear much ab...

For Python package management I use a mixture of pixi, uv and conda depending on the task I'm doing.

I wrote up a long form post about the history of these tools, why each one exists, and why I settled on these choices in my workflow.

jacobtomlinson.dev/posts/2025/p...

18.11.2025 12:56 πŸ‘ 40 πŸ” 21 πŸ’¬ 1 πŸ“Œ 1
Post image

This is terrifying.

"[AI agents] can... infer a researcher's latent hypotheses and produce data that artificially confirms them."

...

"We can no longer trust that survey responses are coming from real people" -@seanjwestwood.bsky.social

18.11.2025 21:03 πŸ‘ 312 πŸ” 121 πŸ’¬ 7 πŸ“Œ 17

From my colleague; paper here dropbox.com/scl/fi/55uok...

18.11.2025 20:10 πŸ‘ 34 πŸ” 14 πŸ’¬ 1 πŸ“Œ 1
Calvin & Hobbes walking along a tree trunk laid across a river as a bridge.

Calvin & Hobbes walking along a tree trunk laid across a river as a bridge.

40 years ago today the world was introduced to a small boy and his best friend. Happy birthday Calvin & Hobbes.

18.11.2025 06:37 πŸ‘ 2447 πŸ” 796 πŸ’¬ 20 πŸ“Œ 80

To borrow another example, taken from the `dbreg` README: github.com/grantmcdermo...

Here I am running a fixed-effects regression on 180 million(!) row parquet dataset... and it completes **< 2 seconds**... on my laptop 🀯

This is powered by @duckdb.org under the hood.

#rstats #econsky

25.08.2025 18:38 πŸ‘ 84 πŸ” 22 πŸ’¬ 3 πŸ“Œ 4
Post image

What it took to get tenure in the Batman Studies Department in the 1960s, and what it takes to get tenure in Batman Studies today...

12.07.2025 02:24 πŸ‘ 782 πŸ” 125 πŸ’¬ 9 πŸ“Œ 8

The shuttle bus

05.05.2025 11:27 πŸ‘ 0 πŸ” 0 πŸ’¬ 0 πŸ“Œ 0

I once wrote to a full professor in Econ for the code - "code was available on request". The professor asked me to code it myself after reading the paper. Lol.

01.05.2025 02:12 πŸ‘ 2 πŸ” 0 πŸ’¬ 0 πŸ“Œ 0
Post image

LLMs are probably trained on a lot of pirated data. This working paper provides interesting evidence.

www.nber.org/papers/w33598

24.03.2025 19:00 πŸ‘ 123 πŸ” 20 πŸ’¬ 3 πŸ“Œ 1
Why DuckDB is my first choice for data processing Why DuckDB has become my go-to tool for data processing, offering simplicity, speed, and powerful features.

New blog: Why DuckDB is my first choice for data processing:
www.robinlinacre.com/recommend_du...

16.03.2025 19:17 πŸ‘ 43 πŸ” 11 πŸ’¬ 0 πŸ“Œ 1

DuckDB for processing, Parquet for publishing. Simple and powerful!

www.robinlinacre.com/parquet_api/

17.03.2025 09:31 πŸ‘ 24 πŸ” 3 πŸ’¬ 0 πŸ“Œ 0
Post image

DuckDB got a local UI. Thanks to our friends at MotherDuck, you can now interact with your DuckDB database through an interactive notebook, running on localhost. Read the announcement blog post at duckdb.org/2025/03/12/d...

12.03.2025 13:18 πŸ‘ 171 πŸ” 47 πŸ’¬ 9 πŸ“Œ 12
Preview
Mutated DNA Restored to Normal in Gene Therapy Advance The small study in patients with a rare disorder that causes liver and lung damage showed the potential for precisely targeted infusions.

Researchers have corrected a disease-causing gene mutation with a single infusion carrying a treatment that precisely targeted the errant gene. The condition affects 100,000 Americans. These genetic treatments were fantasies 30 years ago, but they are now here.

www.nytimes.com/2025/03/10/h...

11.03.2025 02:05 πŸ‘ 40 πŸ” 8 πŸ’¬ 1 πŸ“Œ 0

lol ... Good one!!

11.03.2025 02:13 πŸ‘ 1 πŸ” 0 πŸ’¬ 0 πŸ“Œ 0
Preview
Evidently AI - ML and LLM system design: 500 case studies How do top companies apply AI? A database of 500 case studies from 100+ companies with practical ML use cases, LLM applications, and learnings from designing ML and LLM systems.

This is a pretty cool resource for applied ML: a list of "case studies" sourced from different companies describing problems they face and the methods they've tried to solve them.

Anyone know of something like this specific to geospatial/remote sensing data problems? #MLsky #CCAI #GISchat

08.03.2025 20:00 πŸ‘ 51 πŸ” 9 πŸ’¬ 1 πŸ“Œ 0

Another health breakthrough.

10.03.2025 10:00 πŸ‘ 313 πŸ” 67 πŸ’¬ 12 πŸ“Œ 4
Post image

Unbelievable news.

Pancreatic is one of the deadliest cancers.

New paper shows personalized mRNA vaccines can induce durable T cells that attack pancreatic cancer, with 75% of patients cancer free at three yearsβ€”far, far better than standard of care.

www.nature.com/articles/s41...

27.02.2025 17:03 πŸ‘ 7265 πŸ” 1923 πŸ’¬ 139 πŸ“Œ 315
An image of Aaron Swartz with the following text: "Meta illegally downloaded 80+ TB of books from LibGen, Anna's Archive, and Z-library to train their AI models. In 2010, Aaron Swartz downloaded only 70 GBs of articles from JSTOR (0.0875% of Meta). Faced $1 million in fine and 35 years in jail. Took his own life in 2013.

An image of Aaron Swartz with the following text: "Meta illegally downloaded 80+ TB of books from LibGen, Anna's Archive, and Z-library to train their AI models. In 2010, Aaron Swartz downloaded only 70 GBs of articles from JSTOR (0.0875% of Meta). Faced $1 million in fine and 35 years in jail. Took his own life in 2013.

My heart ached when I was reminded that Aaron Swartz was prosecuted for trying to make academic knowledge more accessible to the public. Meta, meanwhile, is doing it for their own bottom line. I'm going to guess that no one at Meta will be looking at spending 35 years in jail for this.

11.02.2025 23:45 πŸ‘ 2190 πŸ” 943 πŸ’¬ 22 πŸ“Œ 19
Post image

A *2011* grant from NSERC Research Tools and Instruments, for 60000 CAD, to Geoff Hinton (@geoffreyhinton.bsky.social)

For "Large scale machine learning using GPUs"

Everything you're seeing now started from a pittance of cash given by the Canadian government for fundamental research

07.02.2025 17:22 πŸ‘ 93 πŸ” 19 πŸ’¬ 1 πŸ“Œ 1

Can't agree more. (BTW, republicans spending ~ $200 million on anti transgender ads might have had an effect on the narrative)

22.01.2025 12:39 πŸ‘ 0 πŸ” 0 πŸ’¬ 0 πŸ“Œ 0

From my conversations, most seem to have a fear that their kids might get affected and change their gender/sexual orientation.

22.01.2025 10:47 πŸ‘ 2 πŸ” 0 πŸ’¬ 0 πŸ“Œ 1
Change in Paris air quality between 2007 and 2023 using visuals

Change in Paris air quality between 2007 and 2023 using visuals

Air pollution has dropped significantly in #Paris in the last 15 years. Mayor @annehidalgo.bsky.social’s leadership has traded car space for green space, safe bike space, kid space.. and traded pollution for people.

Good trade.

17.01.2025 00:13 πŸ‘ 15263 πŸ” 3592 πŸ’¬ 167 πŸ“Œ 372

Here are the first five sets of slides:

01 Introduction: psantanna.com/DiD/01_Intro...

02 Classical 2x2 setup: psantanna.com/DiD/02_two_b...

03 Clustering issues: psantanna.com/DiD/03_Clust...

04 Functional form: psantanna.com/DiD/04_Funct...

05 Covariates: psantanna.com/DiD/05_Covar...

30.12.2024 05:19 πŸ‘ 732 πŸ” 181 πŸ’¬ 52 πŸ“Œ 15

When he was in the Navy, Jimmy Carter’s submarine was once docked in Bermuda. Some British officers invited only the white crew members to a cocktail party or something like that, and Carter said that if the crewmen of color weren’t invited then nobody was going. This was in the 1950s.

29.12.2024 21:25 πŸ‘ 5890 πŸ” 1070 πŸ’¬ 81 πŸ“Œ 38

Jimmy Carter was the only US President to have lived in public housing.

He lived a life of service and consequence. May he rest in peace.

30.12.2024 00:41 πŸ‘ 400 πŸ” 65 πŸ’¬ 2 πŸ“Œ 5