Cotton production in Mali
Description
This dataset contains annual macro-agricultural and economic data used to analyze the resilience of the Malian cotton sector to climate variability, global price fluctuations, exchange-rate movements, and subsidy policies over the period 1990–2023. The data support a Structural Vector Autoregression (SVAR) analysis examining the dynamic transmission of external and domestic shocks to real cotton producer income per hectare. The dataset includes five core variables: (i) real cotton income per hectare (deflated using Mali’s rural CPI); (ii) annual rainfall as a proxy for climatic conditions; (iii) the Cotlook A Index as a measure of world cotton prices; (iv) the FCFA/USD exchange rate capturing macro-monetary exposure; and (v) input subsidies per hectare reflecting agricultural policy support. All monetary variables are expressed in real terms, and logarithmic transformations are applied where appropriate to facilitate elasticity interpretation and reduce heteroskedasticity. The dataset is designed to enable replication of the empirical results presented in the associated manuscript and to support further research on income dynamics, price transmission, and policy effectiveness in export-oriented agricultural systems in Sub-Saharan Africa.
Files
Steps to reproduce
Download the dataset from the Mendeley Data repository. Import the data into a statistical software environment (Stata). Deflate all nominal monetary variables using Mali’s rural Consumer Price Index (CPI) to obtain real values. Apply natural logarithmic transformations to all variables except precipitation. Conduct unit-root tests (ADF and KPSS) to confirm mixed integration orders (I(0)–I(1)). Estimate a Structural Vector Autoregression (SVAR) model with four lags using short-run (recursive) identification restrictions. Compute impulse response functions (IRFs) and cumulative IRFs to analyze dynamic shock transmission. Perform forecast error variance decomposition (FEVD) to assess the relative contribution of each structural shock. Conduct robustness checks by varying lag length, structural ordering, and subsample periods (1990–2006 and 2007–2023).