Answers are grounded in this paper and cite their sources.
The time periods in most forecasting models of the national economy are quarters, not months or weeks or days. This choice of model frequency is natural since many of the most reliable data series on national economic activity-including the gross domestic product (GDP)-are not published for time periods finer than a quarter. Yet a lot of major national data are published more frequently-employment levels monthly, money measures weekly, and interest rates daily, for example. In fact, many quarterly data series are constructed from finer-time data. So finer-time data should be a source of useful information for anyone interested in analyzing economic activity in the current, incomplete quarter. But how should national forecasters with quarterly models use these finer-time data to improve their quarterly forecasts?
No takes yet. Share an insight, caveat, or question.
Chin et al. (1996) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: