Tools you will need:
- R
- Internet Access
- a Linux PC or a Mac (Windows is currently not supported)
- What you are advised to have: RStudio
Instructions:
1. Open R (if possible, open RStudio) and run the command: install.packages("rover", repos = c("https://lair.functionalfinances.com/repos/shared", "https://lair.functionalfinances.com/repos/cran"))
2. Clone the project https://stash.functionalfinances.com/scm/osr/psychml.git (this is easy in RStudio if you have git installed: File -> New Project -> Version Control -> Git -> use this link and dedicate a folder to this project)
3. If in RStudio, this will automatically open the project. If in pure R, then you will have to set the working directory by hand using setwd command
4. install.packages("lattice", "futile.logger", "mlr")
5. Wait until the installation finishes.
6. Source the random_noise_test.R file to run the random noise test: source("random_noise_test.R")
7. Source the usage.R file to run the analysis: source('usage.R")
(NB: Please note that the scripts have to be run by hand for the different seeds because the party package does not support being called from inside a loop, e.g. from a for loop or an lapply. This is by design and will not be updated according to the authors of the package.)
For researchers who would like to use the scripts for their own research, the annotated scripts are also available [here][1].
[1]: https://osf.io/26kr7/