Arnold Zellner's Contributions to the Structural Econometric Modeling and Time Series Approach (SEMTSA)

• Activity Number: 217
• Type: Invited
• Date/Time: Monday, August 1, 2011 : 2:00 PM to 3:50 PM
• Sponsor: Memorial
• Author(s): Franz C. Palm*+
• Companies: Maastricht University
• Keywords: structural modeling ; time series analysis ; macroeeconomic forecasting ; volatility modeling ; common features

Abstract

In this talk I shall pay tribute to Arnold Zellner for his significant and path-breaking contributions to econometric modeling by reviewing the SEMTSA approach that he put forward almost forty years ago and to which I also had the privilege and pleasure to contribute to by collaborating closely with Arnold. Over the years Arnold continued to develop this approach and applied it to macroeconomic forecasting and modeling. A major feature of this approach is the emphasis on the sequential nature of econometric modeling, whereby coherence requirements are checked such as: the statistical properties of the marginal and conditional pdfs implied by an entertained multivariate model should be in agreement with those of the marginal and conditional pdfs as obtained from empirical analysis of the time series to be modeled. To illustrate the significance of Arnold's contributions, I shall also relate them to recent contributions to SEMTSA, for instance applying it in multivariate volatility modeling and to models which allow for co-movements and other common features of the data.