This repository has been developed for the Panel Discussion on Replication at the 2019 JAR Conference.

It provides the code for the Discretionary Accruals Exploration Platform.

The display estimates Modified Jones and Dechow and Dichev discretionary accruals and presents them with selected firm-year accounting data for a large panel of U.S. incorporated non-financial firms. The methodology loosely follows Hribar and Nichols (JAR, 2007). The discretionary accruals are estimated for all Fama/French 48 industry-year bins with at least 10 observations available.

See run_me.R for the main code and code/pull_wrds_code.R for the code that pulls Compustat and Audit Analytics data from WRDS.

Edit and source run_me.R in a reasonably up-to-date R/RStudio environment to locally regenerate the shiny platform.

The code is based on my 'ExPanDaR' R-package. Feel free to reach out via email or twitter if you happen to have remarks about this project.

Some additional links that might be of interest:

• Explore the Preston Curve (association of national income with life expectancy)

• Explore EPA data on fuel efficiency

• Explore your own data (No worries, your data won't be stored)

• 'ExPanDaR' R package that provides the exploration infrastructure

• R package for Researcher Degrees of Freedom Analysis

• Blog with further info - only partly related

Enjoy!