Predictable Recoveries

Abstract

A random walk with drift is a good univariate representation of US GDP. This paper shows, however, that US economic downturns have been associated with predictable short-term recoveries and with changes in long-term GDP forecasts that are substantially smaller than the initial drop. To detect these predictable changes, it is important to use a multivariate time series model. We discuss reasons why univariate representations can miss key characteristics of the underlying variable such as predictability, especially during recessions.

The Present Value of Corporate Profits

“…Intuitively for the analyst, the growth profile looks 'right' and smooth, unlike forecasting with 'zig-zags'. Estimating an auto-regressive model across survey vintages v with variable long-run growth, Low-order univariate AR or ARMA forecasting models will a have hard time capturing cyclical components on top of integrated processes, see Cai, den Haan, and Pinder (2016). Specifying a trend-cycle state-space model is more involved and can simply be just more complicated for many practitioners.…”