This research has the potential to offer new contributions in the application of statistical methods in agriculture. The aim is to provide a stronger basis for farmers and the government in increasing cassava production in the future based on planting distance and land area. The method used is the Mean Absolute Error (MAE) in the Multiple Linear Regression Algorithm. The result partial test show that planting distance has a real and positive effect on cassava production in Central Buton, as evidenced by the value of T_value= 2.320 > T_table= 1.997, with a significance value of 0.023 < 0.05. Likewise, the partial test results show that land area has a real and positive effect on cassava production, as indicated by the value of T_value= 7.095 > T_table = 1.997, with a significance value of 0.000 < 0.05. Meanwhile, the simultaneous test results show that planting distance and land area have a real and positive effect on cassava production, as indicated by the value of F_value= 56.087 > F_table = 3.986, with a significance value of 0.000 < 0.05. The magnitude of the combined effect, based on the coefficient of determination test, is 62.2%. Thus, the Mean Absolute Error (MAE) method, the average prediction error of the Multiple Linear Regression model obtained obtained very good or acceptable results, namely 111 or around 0.3% of the average error compared to the actual value.
Copyrights © 2026