Zach Flynn
Home Projects Papers Toolshed Presentations & Consulting/Freelance & Blog LinkedIn X/Twitter Github & CVRead my blog at https://goodenoughstatistics.com and/or follow links to individual posts below for my short writing.
Experimentation
- It Takes Too Long To Experiment.
- Experimentation is a reasonable tax.
- Stop the Count! Why Putting A Time Limit On Metrics Is Critical for Fast and Accurate Experiments
- Scale Experiment Decision-Making with Programmatic Decision Rules
- Does It Matter That Online Experiments Interact?
- Instead Of Testing For Sample Ratio Mismatch (SRM), Correct For It
- Losing Is The Value Of Experimentation
- Bounds on Long Run Treatment Effects via d-Surrogacy
- Peeking Not Considered Harmful
- The Best Way To Do Sample-Sizing For An Experiment
- Use Decision Theory to Choose Significance Levels for Experiments
- Making Social Decisions from Experiments: Convincing the Skeptic
- Accelerating Online Experiments that Target Quantile Treatment Effects
Pricing, Economics, and Identification
- Combining Structural Models And Causal Inference Methods
- Which Companies Will Replace Us With AI Robots?
- The Tariff Calculation and Bad Goals
- Parametric estimation vs identification
- Should we compute confidence intervals for partially identified parameters?
- RICE Scoring: an economic model
- Making inference on optimization problems
Statistics and Causal Inference
- In Defense of Statistical Significance
- A simple workhorse method for picking regression specifications
- Outliers
- High Velocity Metrics!
- Monotone Trends for the Difference in Difference Setup
- Trending towards significance
- Central Limit Theorems and the Deeper Naivete
- Estimating Logistic Regressions with Linear Regressions
Personal Blogging (Navel-Gazing)
- Medium and Substack
- Dog Food and Causal Inference
- 18 Years of Wireless Connections
- A 1950s Science Fiction Book About Remote Work
- Know where the time went