Helle Bunzel and I co-taught a PhD elective in Macroeconometrics this semester and there are about four weeks left. One of the more frustrating aspects of the class has been the lack of a really good textbook. Jim Hamilton’s Time Series Analysis (1994) is good but old and is missing a lot of recent material.
The other part of the class that’s hard is teaching students how to think about empirical analysis. Macro research is very model driven, so books like DeJong and Dave’s (2011) Structural Macroeconomics focus more on how to fit particular models to time series data and less on how to decide on an overall approach to modeling. I have mixed feelings about how this influences research, but the point I’m making now is that it doesn’t lead to the sort of textbook that I prefer: Bill Cleveland’s (1993, 1994 resp.) Visualizing Data and Elements of Graphing Data are probably my favorite books ever.
So one approach is to try to expose students to several shorter expository articles. Unfortunately I don’t know of many. These are an incomplete starting point, though:
- Chris Sims (2010), “But Economics is not an experimental science.”
- James Stock (2010), “The other transformation in Econometric practice: robust tools for inference.”
- Both of those articles are in response to Angrist and Pischke (2010), “The Credibility Revolution in Empirical Economics: How Better Research Design Is Taking the Con out of Econometrics.” (These articles are from a symposium on empirical Economics published in the Journal of Economic Perspectives. The other articles in that symposium are interesting as well, but less relevant for Macroeconomics).
Let me know if there are other (recent) papers or books I should add.