Working Paper: NBER ID: w7023
Authors: James H. Stock; Mark W. Watson
Abstract: This paper investigates forecasts of U.S. inflation at the 12-month horizon. The starting point is the conventional unemployment rate Phillips curve, which is examined in a simulated out of sample forecasting framework. Inflation forecasts produced by the Phillips curve generally have been more accurate than forecasts based on other macroeconomic variables, including interest rates, money and commodity prices. These forecasts can however be improved upon using a generalized Phillips curve based on measures of real aggregate activity other than unemployment, especially a new index of aggregate activity based on 61 real economic indicators.
Keywords: inflation forecasting; Phillips curve; macroeconomic indicators
JEL Codes: E31; C32
Edges that are evidenced by causal inference methods are in orange, and the rest are in light blue.
| Cause | Effect |
|---|---|
| Parameters of the Phillips curve (E31) | Statistical instability in the parameters of the Phillips curve (E31) |
| Using alternative measures of aggregate activity (E10) | Smaller mean squared errors in forecasts (C53) |
| Integrating a new index of 61 aggregate activity indicators (C43) | Enhanced forecasting performance (C53) |