This repository powers the analysis for the trade visualizations on the page: Data Dive: How Russia's invasion of Ukraine will impact Africa using bilateral trade from the BACI dataset and global commodity prices from the World Bank.
BACI provides disaggregated bilateral trade data through official trade data reported on COMTRADE. This analysis uses 2017 Harmonized System product nomenclature. The World Bank provides monthly prices for major global commodities in nominal USD.
In order to reproduce the analysis, Python (>=3.10) is needed. Other required packages are
listed in requirements.txt. Additionally, bilateral
trade data is too large to be stored in this repository and needs to be manually
downloaded from BACI.
The repository includes the following sub-folders:
output: contains clean and formatted csv filed that are used to create the visualizations.raw_data: contains raw data used for the analysis and metadata including product and country codes. Manually downloaded files are added to this folder.scripts: scripts for creating the analysis.codes.pycontains grouped HS codes as lists.read_trade_data.pycontains functions to read BACI trade data from CEPII, do some preprocessing and save the data as a feather file.commodities_analysis.pycontains functions to clean and manipulate commodity price data.trade_common.pycontains functions to manipulate the trade data andstory.pycreates the final csv files used to produce the flourish visualisations. Additionally aconfig.pyfile manages file paths to different folders.
The Centre d'Etudes Prospectives et d'Informations Internationales maintains the BACI Dataset. HS17 data file is used for this analysis.
Direct download link: http://www.cepii.fr/DATA_DOWNLOAD/baci/data/BACI_HS17_V202201.zip
Once the zipped file is downloaded the csv filed for 2018, 2019, and 2020 are used.
These files should be moved to the raw_data folder with the naming convention hs17_{year}.csv".