chapter  3
Expectations and the (un)importance of the real-wage feedback channel
Pages 24

In this chapter1 we address a further open issue of the new Keynesian modeling of the macroeconomy that concerns the equivalence of continuous-time or discrete-time modeling or rather period vs. continuous-time analysis. Our considerations start from the empirical fact that while the actual data-generating process (DGP) at the macrolevel, even in the real markets, is by and large a daily one (concerning averages over the day), the corresponding data-collection process (DCP) on the economy-wide goods and labor markets is (due to technological and suitability issues) often at a much lower frequency (on a monthly or quarterly basis). Under the premise that the dynamical properties of both modeling approaches should not depend on the choice of the period length, and taking into account the fact that the behavior of the macroeconomy is in fact of a quasi-continuous-time nature, implies that empirically applicable period macromodels (using annualized data) should be iterated approximately with a step size between 1/365 and 1/52 of a year in order to assure that they generate qualitative results that are equivalent to the ones of their continuous-time analogs. Such empirically applicable period macromodels will then (for example) typically not be able to give rise to chaotic dynamics in one and two dimensions, suggesting that the literature on such chaotic dynamics is of no empirical relevance; see Flaschel and Proaño (2009) for details.