The hottest Causal Inference Substack posts right now

And their main takeaways

Category

Top Science Topics

Mixtape Mailbag #7: What Happens in Difference-in-Differences if Parallel Trends is satisfied but No Anticipation is Violated?

Scott's Substack • 786 implied HN points • 22 Jan 24

🚌 Education Causal Inference Data Analysis Subscribers

In Difference-in-Differences analysis, parallel trends being satisfied is important.
Understanding and considering the assumption of no anticipation is crucial in the analysis.
Losing the assumption of no anticipation can lead to biases in the results.

Scientific Accuracy and Readability

Scott's Substack • 805 implied HN points • 11 Jan 24

🔬 Science Research Causal Inference Methods Data Analysis

Scientific manuscripts should strike a balance between readability and accuracy.
Clearly state the specific target parameters in research questions to drive methods used.
Different methods in scientific research identify different treatment effects; state target parameters upfront to guide method selection.

Labor's Design and the Making of Good Music

Scott's Substack • 687 implied HN points • 18 Jan 24

🔬 Science Causal Inference Music

The paper discusses the importance of 'design' in addressing selection bias in causal inference.
Design focuses on understanding the mechanisms behind treatment assignment in data analysis.
Good writing is crucial in academic work, emphasizing the significance of clear communication in quantitative research.

Workshop announcement: Causal I

Scott's Substack • 334 implied HN points • 12 Jan 24

🚌 Education Causal Inference Statistics Teaching Learning

Workshop announcement for Causal Inference I starting on February 3rd.
Key topics covered in the workshop include potential outcomes and selection bias.
The importance of letting go of anger, bitterness, and seeking human connection in New Year's resolutions.

Difference-in-Differences: No Anticipation with Parallel Trends (with a simulation and equations)

Scott's Substack • 117 implied HN points • 31 Jan 24

🔬 Science Causal Inference Statistics Simulation Data Analysis

No anticipation means the baseline period is equal to Y(0) not Y(1)
Difference-in-differences coefficient equals ATT in the post period for the treatment group plus parallel trends bias minus ATT in the incorrectly specified baseline period
Difference-in-differences always requires three assumptions to point identify the ATT: SUTVA, Parallel trends, and No Anticipation

Get a weekly roundup of the best Substack posts, by hacker news affinity:

Workshop announcement: $95 price for non-tenure track or professors with higher teaching loads

Scott's Substack • 117 implied HN points • 24 Jan 24

🚌 Education Workshop Teaching Causal Inference Pricing Learning

Workshop offering discounted price of $95 for non-tenure track professors or those with high teaching loads
Workshop covers topics like potential outcomes model, unconfoundedness, and instrumental variables
Teaching style focuses on comprehension, confidence, and competency in applying causal inference methods

Workshop announcement: Causal Inference 1

Scott's Substack • 98 implied HN points • 29 Jan 24

🚌 Education Workshop Causal Inference

Workshop on Causal Inference is happening over two weekends in February 2024
Discounts are available for different categories of participants based on their affiliation and location
Emphasis is on employer reimbursement for workshop fees

Mixtape Mailbag #8: Continuous Triple Differences

Scott's Substack • 39 implied HN points • 05 Feb 24

🔬 Science Causal Inference Data Analysis Research Methods Econometrics

Triple difference design can be used with continuous treatment by defining the parameters based on dosage levels.
When treatment is continuous, the target parameter shifts from average treatment effect to average causal response function.
Continuous treatments require careful definition of parameters to compare different dosages along a treatment curve.

[🔥Hot Takes] Deep learning outperforms linear regression for causal inference and tabular data??

Data Science Daily • 0 implied HN points • 02 Mar 23

🕹 Technology Data science Deep Learning Causal Inference Machine Learning Research

Deep learning can outperform linear regression for causal inference in tabular data.
Different perspectives exist in the debate between deep learning and traditional models like XGBoost.
The study suggests that deep learning models like CNN, DNN, and CNN-LSTM may offer better performance in certain scenarios.