Scratchpad – Daniel Molitor

Daniel Molitor

Research
CV
Scratchpad

Categories

All (5)

Adaptive Experiments (1)

Anytime-valid Inference (3)

AWS (1)

Bandits (2)

Conformal Prediction (1)

Conjoints (1)

Docker (1)

Shiny (1)

Anytime-valid Inference on the AMCE in Conjoint Experiments

Conjoints

Anytime-valid Inference

An empirical example based on Causal Inference in Conjoint Analysis: Understanding Multidimensional Choices via Stated Preference Experiments by Hainmueller, Hopkins, and Yamamoto.

Anytime-Valid Regression Adjusted Causal Inference

Anytime-valid Inference

Investigating the methods described in Anytime-Valid Linear Models and Regression Adjusted Causal Inference in Randomized Experiments by Lindon, et al. via simulations.

Robust Adaptive Experiments

Adaptive Experiments

Anytime-valid Inference

Bandits

Exploring the MAD design (Liang and Bojinov, 2024) and extending it to balance anytime-valid inference, reward maximization, and statistical power in adaptive experiments.

Conformal Prediction Beyond Exchangeability

Conformal Prediction

Replicating empirical examples from Conformal Prediction Beyond Exchangeability by Barber, et al.

Building and Deploying Adaptive Experiments with Shiny

AWS

Bandits

Docker

Shiny

A case study on how Shiny + Docker + AWS can help researchers build and deploy adaptive experiments and custom surveys.

No matching items

Built with Quarto