This preview shows pages 1–3. Sign up to view the full content.
This preview has intentionally blurred sections. Sign up to view the full version.
View Full Document
Unformatted text preview: MIT OpenCourseWare http://ocw.mit.edu 14.384 Time Series Analysis Fall 2008 For information about citing these materials or our Terms of Use, visit: http://ocw.mit.edu/terms . Goals &amp; Assumptions 1 14.384 Time Series Analysis, Fall 2007 Professor Anna Mikusheva Paul Schrimpf, scribe October 4, 2007 Lecture 9 Structural VARs Goals &amp; Assumptions This lecture is about applied Macroeconomics. This is not a course about macroeconomics, so we wont have too much to say about the correctness of various macroeconomic assumptions. However, we will try to clearly separate econometric issues from macroeconomic ones. History Sims (1980) Macroeconomics and Reality introduced VARs. It is largely a philosophical paper. The com mon practice at the time was to estimate very large macromodels with many equations and many restrictions. Sims argued that these restrictions were often unrealistic. Sims introduced VARs as an alternative. 1. Estimate by OLS A ( L ) y t = e t , a V AR ( p ) 2. Invert VAR to get MA ( ), y t = C ( L ) e t 3. Identification: find matrix D such that Du t = e t where u t are orthonormal, i.e. Eu t u t = I t , and serially uncorrelated 4. Compute impulse responses from y C t = ( L ) u t , where C j = DC j Goals Causal relation eg. how does money supply affect output? Policy analysis eg. how should fed adjust interest rate? Test economic theories eg. test RBC against NeoKeynesian models as in problem set labor productivity Example 1 . y = . RBC implies that a permanent shock to labor productivity labor hours should cause hours to go up. NeoKeynesian implies the opposite....
View
Full
Document
 '09
 Geurrieri

Click to edit the document details