A tool to explore the effect of the Iberian exception mechanism on inflation. It can be used to replicate the results of Haro Ruiz, M, Schult, C., Wunder, C. (2024). The effects of the Iberian Exception mechanism on wholesale electricity prices and consumer inflation. A synthetic-controls approach.
The are two options to execute ibex; either locally or via a Docker container.
To run the tool locally, clone the repository and make sure that all the necessary R packages are installed. To do so, you may run install_packages.R before executing main.R. This option is likely to result in issues due to conflicting R versions and package versions.
As an alternative, ibex can be deployed in a Docker container. To do so, you must have Docker installed on your computer.
Begin by cloning this repository from the command line:
git clone https://github.com/mharoruiz/ibex && \
cd ibexMake sure your working directory points to the cloned repository. Depending on your command line, you can check this with pwd or cd.
Next, run a Docker container from an image that includes all the necessary dependencies:
docker run --rm \
-p 8787:8787 \
-e DISABLE_AUTH=true \
-v "$(pwd):/home/rstudio/ibex" \
-v /home/rstudio/ibex/renv \
mharoruiz/ibex:0.1The above command creates and runs a new container (docker run) from the image mharoruiz/ibex:0.1, and mounts the project files (-v) from your local directory as source ($(pwd)) to the container directory as destination (/home/rstudio/ibex).
Troubleshooting
Depending on the command line in your computer, you may have to use %cd% instead of $(pwd). If none of these methods work, you can substitute the source with the absolute path to the cloned repository in your local machine (something like /Users/user/ibex or C:\Users\user\ibex).
Another source for potential errors are the \ at the end of every line in the docker run command above. These allow the execution of multiline and make the code block above easier to read than a single, long line of code. Depending on your command line, you may have to use ^ instead of \. However, you can also remove the linebreaks and execute the command as a single line.
The first time you execute docker run successfully, the command line should inform you that the image is being pulled from a Docker repository with the following message:
Unable to find image 'mharoruiz/ibex:0.1' locally
0.1: Pulling from mharoruiz/ibex
This process may take some time. However, subsequent executions of docker run will use a locally saved image to run the container. When this happens, your command line should print messages indicating that the container is being initialized. You can now access the container by pointing your browser to localhost:8787. This will open a ready-to-use instance of RStudio.
Open ibex/ibex.Rproj in the RStudio session to load the project dependencies. The following message should appear in your R console:
- Project '~/ibex' loaded. [renv 1.0.2]The script ibex/main.R replicates the results in Haro Ruiz, M., Schult, C. & Wunder, C. (2024). Execute the script by pressing Cmd/Ctrl + Shift + S or using the following command:
source("~/ibex/main.R")Once main.R has successfully run, the figures and tables will be accessible as variables in your R environment, i.e. fig_1. Additionally, if constant SAVE_ANALYSIS=TRUE, the figures will be saved to 04_analysis/ as .png files.
The runtime of the main.R depends on constant PRC_STEP, which is defined in line 22 of the script and determines the precision of the confidence intervals for the treatment effect. By default, PRC_STEP=.1, which allows for a relatively quick execution. Note that the results presented in the paper were obtained with PRC_STEP=.001 (These results are saved to 03_results/sc_series_001.csv).