Learning and time-varying macroeconomic volatility

Abstract

a b s t r a c tThis paper presents a DSGE model in which agents' learning about the economy can endogenously generate time-varying macroeconomic volatility. Economic agents use simple models to form expectations and need to learn the relevant parameters. Their gain coefficient is endogenous and is adjusted according to past forecast errors.The model is estimated using likelihood-based Bayesian methods. The endogenous gain is jointly estimated with the structural parameters of the system.The estimation results show that private agents appear to have often switched to constant-gain learning, with a high constant gain, during most of the 1970s and until the early 1980s, while reverting to a decreasing gain later on. As a result, the model can generate a pattern of volatility, which is increasing in the 1970s and falling in the second half of the sample, with a decline that can roughly match the magnitude of the so-called "Great Moderation" in the 1984-2007 period. The paper also documents how a failure to incorporate learning into the estimation may lead econometricians to spuriously find time-varying volatility in the exogenous shocks, even when these have constant variance by construction.

On the initialization of adaptive learning in macroeconomic models

“…Depending on the weighting scheme and the prior estimates, there are two main variations of this method in the learning literature: the OLS-based (e.g., Williams, 2003;Orphanides and Williams, 2005a;Sargent et al, 2006) and the WLS-based (e.g., Primiceri, 2006;Milani, 2007Milani, , 2008Milani, , 2011Milani, , 2014Huang et al, 2009;Chevillon et al, 2010;Eusepi and Preston, 2011;Lubik and Matthes, 2014) initials.…”

Empirical calibration of adaptive learning

“…The constant gain specification has been in the spotlight of most applied research since Sargent (1999), given its tracking capabilities and its suitability for time-varying environments. More recently, alternative approaches of time-varying gains also have been proposed to deal with relatively stable systems that occasionally undergo structural breaks (see Marcet and Nicolini, 2003;Kostyshyna, 2012;Milani, 2014); one key issue in the introduction of time-varying gains is to find an appropriate statistical foundation for the gain adjustment mechanism.…”

“…Moreover, we explore different elements of interest for applied research, such contribution in the applied literature is Sargent (1999), with a study on the role of learning by the monetary authority in the evolution of US inflation rates, followed by Marcet and Nicolini (2003), who showed how learning can help explain the recurrence of hyperinflation episodes observed in several South-American countries during the 1980's. The literature that followed then unfolded into two main strands: policy-oriented studies of particular episodes of macroeconomic turmoil (mainly the US Great Inflation period during the 1970's; see, e.g., Bullard and Eusepi, 2005;Orphanides and Williams, 2005;Primiceri, 2006;Sargent et al, 2006;Milani, 2008); and business cycles modeling studies (Williams, 2003;Huang et al, 2009;Chevillon et al, 2010;Eusepi and Preston, 2011;Slobodyan and Wouters, 2012;Milani, 2007Milani, , 2011Milani, , 2014Ormeño and Molnár, 2015).…”

The Impact of Attention to News about Tax Changes on the Stock Market

“…Psychological factors have been invoked by many researchers; Evans and Honkapohja (2001), Evans and Honkapohja (2003), Milani (2014) and CarcelesPoveda and Giannitsarou (2008). According to the topic of the paper, the literature has witnessed many attempts to measure investor attention.…”