Stochastic Prediction in Dynamic Nonlinear Econometric Systems

• Main Authors: Mariano, Roberto S., Brown, B.W.
• Format: text
• Language: English
• Published: Institutional Knowledge at Singapore Management University 1985
• Subjects: Economics
• Online Access: https://ink.library.smu.edu.sg/soe_research/65
• Institution: Singapore Management University

This paper considers the large-sample asymptotic behavior of predictors in dynamic nonlinear econometric models. The analytical results summarized in this paper document potential deficiencies in the common practice of forecasting through deterministic simulations of the nonlinear model. For asymptotic prediction efficiency, alternative forecasting procedures based on stochastic simulations of the model are analyzed. In particular, the paper focuses on stochastic simulations based on calculated residuals to develop more robust forecasting procedures in dynamic nonlinear systems.