Quantifying the quality of macroeconomic variables

Methods to quantify the quality of a macroeconomic statistical time series are presented. The integrated measures are based on a combination of how predictable the series is and how much its statistics need to be revised. An ''information window'' based on signal-to-noise ratios provides a snapshot of the quality. A formulation of information in terms of entropy is also considered. Our approach allows testing of whether a forecast or a preliminary value is informative. Concavity and monotonic convergence of information accrual are discussed.