Figure 7
A horizontal bar graph titled 'Random Forest Feature Importance Rankings' compares the importance of different features in a Random Forest model. The x-axis represents 'Feature Importance' with values ranging from 0.0 to 0.5. The y-axis lists the features: rolling_mean_3, price_change, lag_1, lag_2, lag_3, year, month, commodity_Rice (local), admin1_Yobe, commodity_Groundnuts (shelled), admin1_Borno, commodity_Millet, commodity_Cowpeas (brown), commodity_Yam, and admin1_Kaduna. The bars are colored blue. The rolling_mean_3 feature has the highest importance score at approximately 0.488, followed by price_change at 0.118, lag_1 at 0.117, lag_2 at 0.062, and lag_3 at 0.052. The remaining features have significantly lower importance scores, all below 0.05.Random forest feature importance rankings showing the dominant contribution of temporal features to agricultural price prediction. (Source: Authors’ own work)