[gov3009-l] James Stock on ‘Forecasting in Dynamic Factor Models’

The Applied Statistics Workshop is proud to present James Stock, Chair of the Economics Department, as he presents, "Forecasting in Dynamic Factor Models Subject to Structural Instability". James has provided the following abstract: Dynamic factor models (DFMs) express the comovements of time series at leads and lags in terms of a small number of latent factors. In macroeconomic applications, the latent factors can be thought of as theoretical constructs (income) that are linked to specific measurements (GDP). The large body of work on DFMs in macroeconomics assumes a stable structure. This paper develops time-varying DFMs and uses implications of time-varying DFMS to shed light on some ongoing macro puzzles such as the Great Moderation and the breakdown of the backward-looking Phillips curve. The workshop will meet at 12-noon in room N-354, CGIS-Knafel. And a light lunch will be served. Hope you can make it, Justin Grimmer