SUPR
Experimental designs for causal effects
Dnr:

NAISS 2024/5-246

Type:

NAISS Medium Compute

Principal Investigator:

Fredrik Sävje

Affiliation:

Uppsala universitet

Start Date:

2024-07-01

End Date:

2025-07-01

Primary Classification:

10106: Probability Theory and Statistics

Secondary Classification:

50201: Economics

Tertiary Classification:

10201: Computer Sciences

Allocation

Abstract

Randomized experiments assign causal factors (or treatments) under study at random to the units in an experiment sample. This facilitates investigation of causal effects without strong assumptions, which has made the method the gold standard in causal inference in many fields in the social and medical sciences. This project aims to investigate what probability distributions (typically called experimental designs) to use when drawing the randomized treatments. The cluster will be used primarily for Monte Carlo investigations of various designs.