HeyJaredScott's Mixtape Substack
Scott’s Mixtape is dedicated to teaching causal inference, econometrics, and applied empirical research with a focus on how AI agents are transforming how researchers work. Practical tutorials, live research, and documentation of what works and doesn’t.
Part of The Mixtape with Scott. Follow the publication for all its platforms.
- Indexed issues, last 90 days
- 14
- Latest publication
- Sep 28, 2026
- Audience
- Checking…
- Earliest in this view
- Aug 17, 2026
Latest issues
Nobel prize predictions: for the wages of nations (opens the original)
Read excerpt
I like to open my predictions about the Nobel prize in economics by making a joke about how someone on the internet will correct me and say it’s not a real Nobel. But this time, I’ll just link to an earlier substack where I made that joke while I simply announce that it’s the time of year and I would like submit my prediction. I’ve had mostly a good record at it, usually winning my predictions, more or less. I basically called the Card, Angrist and Imbens award. Then I messed up. I guess to myse
CodeChella Mexico City! (opens the original)
Read excerpt
A few weeks ago I posted about the upcoming CodeChella conference we’re hosting in Mexico City. I will likely be posting about it regularly to try and hype it up. The sooner you sign up, the sooner you’re guaranteed a slot! Let me tell you about it, but before I tell you about it, I also wanted to share something.<a class="image-link image2 is-viewable-img" href="https://substackcdn.com/image/fetch/$s_!mt2c!,f_auto,q_auto:good,fl_progressive:steep/https%3A%2F%2Fsubstack-post-media.s3.amazonaws.c
Parallel Trends is a Debiasing Method (opens the original)
Read excerpt
Let’s write down a simple regression equation. Let Y be the outcome, D be a dummy and epsilon is an error term. The interpretation of the coefficient on D is that it is the simple difference in mean outcomes. Take the average of the outcome for the treatment group (D=1) and subtract the average of the outcome for the control group (D=0), and that is the same as running that regression and looking at delta hat.In my book, I decompose a simple comparison in means into treatment effects and selecti
Should I combine two comparison groups from two distinct regimes in my diff-in-diff? (opens the original)
Read excerpt
I was reading today that some Massachusetts counties are discontinuing their contracts with Flock, the AI technology firm that provides surveillance products, particularly of vehicles (and license plates). This was the line.“A growing number of Massachusetts communities are reconsidering their relationships with Flock Safety, the company behind the controversial Flock automated license plate reader cameras.Some Massachusetts towns have ended their c
Connectors and My Harness Idea (opens the original)
Read excerpt
I have been using the connectors in Claude Code fairly regularly now. Connectors seem to be the mechanism by which I can imagine being able to do two things.Being able to use Claude Code from within the claude app on my phoneBeing able to extend Claude Code into cloud services, not just my local computerSo what are connectors and sorts of use cases am I seeing? Let me first show you where it is located as I’m not sure everyone has seen it, or is using it. Here’s the screenshots to it on the Mac
Publishing over time
Last 90 days. Choose a month to open its work.
Recurring subjects
Named in the text we hold. One piece can cover several.
Audience
No verified audience measurement yet.
About this data
Counts cover the work we have indexed. Tone needs enough text and a confident classification. Excerpts and episode notes are not full articles or transcripts.
Identity or attribution wrong? Suggest a correction.