Harmonized Multi-Source Dataset of Agricultural Commodity Prices, Meteorological Variations, and Macroeconomic Indicators for Mali
Description
This dataset provides a comprehensive, multi-variable panel combining agricultural market commodity prices, meteorological factors, and macroeconomic indicators across major administrative regions and markets in Mali. The primary objective of this compiled dataset is to support predictive modeling, econometric forecasting, machine learning applications, and vulnerability assessments concerning cereal price volatility and agricultural market integration in the West African Sahel. The dataset compiles information from multiple authoritative open sources: - Agricultural Commodity Prices: Cereal and staple crop price dynamics across local retail and wholesale markets, originating from the World Food Programme Vulnerability Analysis and Mapping (WFP VAM) database. - Climate and Weather Variables: Historical meteorological indicators (such as precipitation, 2-meter surface temperatures, and related agro-climatic metrics) extracted and aggregated via the Open-Meteo historical weather platform. - Macroeconomic Indicators: Global crude oil price benchmarks and imported inflation indices reflecting broader economic fluctuations impacting domestic food supply chains. The provided tabular file (df_final.csv) contains temporally aligned, geographically indexed, and pre-cleaned records suitable for immediate statistical analysis, time-series forecasting, and econometric modeling without additional feature merging requirements.
Files
Steps to reproduce
The harmonized data table was constructed following a four-stage pipeline: 1. Data Acquisition: - Market-level price series for staple commodities were retrieved from the WFP VAM price monitoring portal for monitored markets across Mali. - Meteorological data were extracted programmatically using the Open-Meteo Historical Weather API by specifying geographical bounding coordinates matching the monitored market clusters. - International crude oil spot prices and national import-driven inflation indicators were collected from public financial and macroeconomic repositories. 2. Temporal and Spatial Alignment: - Temporal frequencies were standardized and resampled to a consistent tracking interval. - Market locations and administrative divisions were normalized using standard ISO and administrative boundary conventions to ensure accurate spatial joining. 3. Feature Cleaning and Imputation: - Outliers resulting from localized transmission errors were validated against surrounding regional market trends. - Missing records in meteorological indicators were cross-referenced with spatial interpolation techniques from neighboring grid points. 4. Integration: - All feature vectors were joined on composite keys [Date, Market/Region ID] to form the single harmonized table (df_final.csv).