Working Papers,
Lund University, Department of Economics

Abstract: You want to run an experiment where you anticipate finding a treatment effect. How large should your sample size be to have a reasonable chance of detecting significant results? In this manuscript, I present and explain the Stata code I use to address this question. The code uses simulations to conduct power analyses, offering a flexible alternative to the commonly used analytical tools. Unlike traditional methods, this approach can accommodate any experimental design and statistical test that Stata supports. The code is straightforward, user-friendly, and can be used effectively with minimal coding experience.

Keywords: Power analysis; Simulations; Stata

JEL-codes: C15; C90

Language: English

15 pages, August 21, 2024

Full text files

WP24_4  Full text

Download statistics

• Lund University, Department of Economics
• Home page for this series

Questions (including download problems) about the papers in this series should be directed to Iker Arregui Alegria (wp-editor@nek.lu.se)
Report other problems with accessing this service to Sune Karlsson (sune.karlsson@oru.se).

RePEc:hhs:lunewp:2024_004This page generated on 2024-09-13 22:16:10.