Constructing fan charts from the ragged edge of SPF forecasts

We develop a model that permits the estimation of term structure both expectations and forecast uncertainty for application to professional forecasts such as Survey Professional Forecasters (SPF). Our approach exactly replicates given data set predictions from SPF (or similar source) without measurement error. captures fixed horizon fixed-event forecasts, can accommodate changes in maximal available SPF. The casts decomposition multi-period errors into sequence updates may be partially unobserved, resulting multivariate unobserved components model. In our empirical analysis, we provide quarterly structures bands. preferred specification features stochastic volatility updates, which improves performance yields estimates vary over time. conclude by constructing SPF-based fan charts calendar-year like those published Federal Reserve. Replication files are at https://github.com/elmarmertens/ClarkGanicsMertensSPFfancharts.