Randomized trial predicts agricultural investment success in semi-arid regions, suggesting guidance for policy decisions.
Key Points
This research aims to develop a predictive model for agricultural investment success in semi-arid regions of Algeria, focusing on cereal farming systems.
Utilized survey data from 198 farms in Setif, Algeria.
Built seven composite asset dimensions and established farm typologies using multivariate analysis.
Developed a binary logistic regression model to predict agricultural investment success.
Achieved 79.8% overall classification accuracy with a R² of 0.358 and R² of 0.486.
Technical Asset was the strongest positive predictor (B = 0.728, OR = 2.072, p = 0.002).
Identified negative predictors such as Diversity of Farming and Connectivity, indicating complex investment dynamics.