Reply to Discussion of “Bayesian forecasting of multivariate time series: scalability, structure uncertainty and decisions”
- 17 Downloads
I am most grateful to the invited discussants, Professor Chris Glynn and Dr. Jouchi Nakajima, for their thoughtful and constructive comments and questions. Their discussion contributions speak clearly to some of the key areas of advance in Bayesian forecasting and time series modeling reviewed in the paper, and critically address important areas of “Challenges and Opportunities” with some new suggestions and connections. My responses here speak directly to their specific comments and questions. I hope and expect that this conversation will additionally contribute to promoting new research developments in dynamic models for increasingly complex and challenging problems in multivariate time series analysis and forecasting—and the broader fields of statistical modeling and decision analysis—in the Akaike tradition.
The discussants focus primarily on issues of model structure specification and learning in dynamic graphical models. These issues raise hard questions in multivariate models...
- Giannone, D., Lenza, M., Primiceri, G. E. (2018). Economic predictions with big data: The illusion of sparsity. Federal Reserve Board of New York, Staff Report No. 847.Google Scholar
- Irie, K. (2019). Bayesian dynamic fused LASSO. Technical Report, Department of Economics, The University of Tokyo, arXiv:1905.12275.
- Kimura, T., Nakajima, J. (2016). Identifying conventional and unconventional monetary policy shocks: A latent threshold approach. The BE Journal of Macroeconomics, 16, 277–300.Google Scholar
- Lavine, I., Lindon, M., West, M. (2019). Adaptive variable selection for sequential prediction in multivariate dynamic models, arXiv:1906.06580.
- Sims, C. A. (2012). Statistical modeling of monetary policy and its effects. American Economic Review, 102, 1187–1205, also: Nobel Prize in Economics documents, Nobel Prize Committee, 2011.Google Scholar