Browsing by Author "Perez, SJ"

Now showing 1 - 4 of 4
  • Thumbnail Image
    ItemOpen Access
    A bootstrap method for identifying and evaluating a structural vector autoregression
    (Oxford Bulletin of Economics and Statistics, 2008-08-01) Demiralp, S; Hoover, KD; Perez, SJ
    Graph-theoretic methods of causal search based on the ideas of Pearl (2000), Spirtes et al. (2000), and others have been applied by a number of researchers to economic data, particularly by Swanson and Granger (1997) to the problem of finding a data-based contemporaneous causal order for the structural vector autoregression, rather than, as is typically done, assuming a weakly justified Choleski order. Demiralp and Hoover (2003) provided Monte Carlo evidence that such methods were effective, provided that signal strengths were sufficiently high. Unfortunately, in applications to actual data, such Monte Carlo simulations are of limited value, as the causal structure of the true data-generating process is necessarily unknown. In this paper, we present a bootstrap procedure that can be applied to actual data (i.e. without knowledge of the true causal structure). We show with an applied example and a simulation study that the procedure is an effective tool for assessing our confidence in causal orders identified by graph-theoretic search algorithms. © 2008. Blackwell Publishing Ltd and the Department of Economics, University of Oxford.
  • Thumbnail Image
    ItemOpen Access
    Empirical Identification of the Vector Autoregression: The Causes and Effects of U.S. M2
    (2008) Hoover, KD; Demiralp, S; Perez, SJ
    The M2 monetary aggregate is monitored by the Federal Reserve, using a broad brush theoretical analysis and an informal empirical analysis. This paper illustrates empirical identification of an eleven-variable system, in which M2 and the factors that the Fed regards as causes and effects are captured in a vector autogregression. Taking account of cointegration, the methodology combines recent developments in graph-theoretical causal search algorithms with a general-to-specific search algorithm to identify a fully specified structural vector autoregression (SVAR). The SVAR is used to examine the causes and effects of M2 in a variety of ways. We conclude that, while the Fed has rightly identified a number of special factors that influence M2 and while M2 detectably affects other important variables, there is 1) little support for the core quantity-theoretic approach to M2 used by the Fed; and 2) M2 is a trivial linkage in the transmission mechanism from monetary policy to real output and inflation.
  • Thumbnail Image
    ItemOpen Access
    Post hoc ergo propter once more an evaluation of 'does monetary policy matter?' in the spirit of James Tobin
    (Journal of Monetary Economics, 1994-01-01) Hoover, KD; Perez, SJ
    Christina and David Romer's paper 'Does Monetary Policy Matter?' advocates the so-called 'narrative' approach to causal inference. We demonstrate that this method will not sustain causal inference. First, it is impossible to distinguish monetary shocks from oil shocks as causes of recessions. Second, a world in which the Fed only announces intentions to act cannot be distinguished from one in which it in fact acts. Third, the techniques of dynamic simulation used in the Romers' study are inappropriate and quantitatively misleading. And, finally, their approach provides no basis for establishing causal asymmetry. © 1994.
  • Thumbnail Image
    ItemOpen Access
    Truth and robustness in cross-country growth regressions
    (Oxford Bulletin of Economics and Statistics, 2004-12-01) Hoover, KD; Perez, SJ
    We re-examine studies of cross-country growth regressions by Levine and Renelt (American Economic Review, Vol. 82, 1992, pp. 942-963) and Sala-i-Martin (American Economic Review, Vol. 87, 1997a, pp. 178-183; Economics Department, Columbia, University, 1997b). In a realistic Monte Carlo experiment, their variants of Edward Leamer's extreme-bounds analysis are compared with a cross-sectional version of the general-to-specific search methodology associated with the LSE approach to econometrics. Levine and Renelt's method has low size and low power, while Sala-i-Martin's method has high size and high power. The general-to-specific methodology is shown to have a near nominal size and high power. Sala-i-Martin's method and the general-to-specific method are then applied to the actual data from Sala-i-Martin's original study.