We examine the role of generalized constant gain stochastic gradient (SGCG) learning in generating large deviations of an endogenous variable from its rational expectations value. We show analytically that these large deviations can occur with a frequency associated with a fat tailed...

We use a continuous-time approximation approach to analyze dynamics of a model where government adaptively learns the Phillips curve while running monetary policy (Phellps problem). This approach is based on approximating the discrete-time dynamics with learning by a limiting continuous-time...

In this paper we provide explicit characterization of the escape dynamics for the Phellps problem of government controlling inflation with adaptive learning of the approximate Phillips curve, alternative to the one considered by Cho, Williams and Sargent (2002). Our approach is based on...

In this paper we provide an explicit characterization of the escape dynamics for the Phellps problem of government controlling inflation with adaptive learning of the approximate Phillips curve, alternative to the one considered by Cho, Williams and Sargent (2002) (CWS in sequel). Our approach...

In this paper we provide an explicit characterization of the escape dynamics for the Phellps problem of government controlling inflation with adaptive learning of the approximate Phillips curve, alternative to the one considered by Cho, Williams and Sargent (2002) (CWS in sequel). Our approach...

We document an unpleasant feature of Epstein-Zin preferences in a stylized model economy of the long-run risk type now widespread in Asset Pricing: Agents with preference parameters commonly described as indicating a "preference for early resolution of uncertainty" achieve higher utility levels...

We study information processing in a simple endowment economy where the mean consumption growth rate are governed by a hidden state variable and agents have recursive preferences. We show that for typical parameter values, there is a strong incentive to commit to ignoring future information on...

Strong consistency of least squares estimators of the slope parameter in simple linear regression models is established for predetermined stochastic regressors. The main result covers a class of models which falls outside the applicability of what is presently available in the literature. An...