Analyzing Saudi Potato Demand: A Comparative Evaluation of AIDS, QUAIDS, and Rotterdam Models

This study evaluates and compares the empirical performance of three demand system models―Almost Ideal Demand System (AIDS), Quadratic AIDS (QUAIDS), and Rotterdam―in estimating price, income, and cross-price elasticities for potato demand in Saudi Arabia. Given the volatile nature of emerging-market data, the study aims to identify the most suitable model for accurately capturing consumer behavior in the Saudi potato market. The empirical analysis highlights the strengths and limitations of each model in handling price and income fluctuations. While the AIDS and QUAIDS models proved theoretically attractive, both exhibited estimation challenges in this context, including statistically insignificant coefficients, violations of homogeneity and symmetry, and unstable elasticity results. QUAIDS, though capable of capturing nonlinearities, suffered from convergence issues and multicollinearity under volatile conditions. These limitations prompted the adoption of the Rotterdam model, which provided a more stable and theoretically consistent forecast. The study also implements homogeneity and symmetry restrictions within Rotterdam’s framework (RMLE) to enhance model reliability. Diagnostic tests, including Wald tests for coefficient significance and Durbin-Watson tests for residual autocorrelation, confirm the robustness of the findings. In addition, the analysis highlights the practical significance of elasticity estimation for policy design and food security planning. The results offer valuable insights for policymakers and market analysts to understand demand elasticity and improve forecasting in Saudi Arabia’s potato sector.