The Econometrics Journal published a Special Issue on the Methodology and Applications of Structural Dynamic Models and Machine Learning, edited by former Co-editor John Rust and guest editors Fedor Iskhakov and Bertel Schjerning. This Special Issue follows from the Second Conference on Structural Dynamic Models, which focused on the use of machine learning and artificial intelligence to facilitate the solution and estimation of dynamic structural models.Fedor, John, and Bertel’s editorial introduces this exciting topic and the Special Issue’s four papers. Mitsuru Igami’s lead article highlights the close relation between AI and structural dynamic econometrics, using computer algorithms for playing board games as examples. Jeppe Druedahl and Anders Munk-Nielsen show how machine learning can be used to flexibly estimate income processes, a key input to structural life-cycle models. Michael Keane and Timothy Neal compare deep neural networks and econometric panel data methods to predict the effect of climate change on crop yields. Finally, Fedor, John, and Bertel themselves conclude with a comprehensive and thought-provoking discussion of the contrasts and possible synergies between structural econometrics and machine learning.
