Forecasting Inflation

A note on the representative adaptive learning algorithm

“…These results for inflation also seem to be consistent to the recent findings of Faust and Wright (2013) showing that simpler forecasting methods tend to outperform more complicated model-based forecasts of inflation. Taking the surveys as a reference, nevertheless, pointed to an overall dominance of the LS in providing forecasts closer to those representing actual agents forecasts.…”

Inside the Crystal Ball: New Approaches to Predicting the Gasoline Price at the Pump

“…Table 8. The pooled forecasts are averaged by quarter, if required, and are inflated using real-time inflation forecasts constructed as in Faust and Wright (2013). The benchmark is the monthly no-change forecast for the nominal price of gasoline.…”

“…A natural question is how accurate these forecasts are and whether they may be improved upon by using the P5 forecast proposed in this paper. We construct quarterly nominal P5 forecasts by averaging the monthly pooled forecasts of the real price by quarter, as in Figure 4, and inflating these forecasts based on the method of Faust and Wright (2013). The benchmark is the monthly nominal no-change forecast.…”

“…These forecasts are converted to nominal terms using real-time inflation forecasts constructed analogously to Faust and Wright (2013). The real price of gasoline is expressed in June 2014 dollars.…”

Testing for Differences in Path Forecast Accuracy: Forecast-Error Dynamics Matter

“…Recent work by Faust and Wright (2013) and Schorfheide and Song (2015) shows that improvements in the nowcasts can lead to improvements in the forecasts. However, the DSGE and Greenbook nowcasts have very different properties which may not result in forecast improvements.…”