We evaluate the performance of inflation forecasts backed out from the nominal and real yield curves in the United Kingdom.

We use vector autoregressions of the Nelson–Siegel factors to forecast inflation.

A model for the one-day break-even inflation including a real Cochrane-Piazzesi factor outperforms other models.

We quantify the parameter uncertainty, showing that inflation forecasts may be subject to considerable parameter uncertainty, in particular after the Lehman Brothers event.