All Times ET
Program is Subject to Change
TBD
A New Look at Signal Extraction for Manufacturers’ Shipments, Inventories, and Orders (M3) Survey (308027)
James Livsey, US Census Bureau*Colt Viehdorfer, US Census Bureau
Keywords: Seasonal adjustment, signal extraction, multivariate, SEATS, X-11
Signal extraction, specifically seasonal adjustment, is ubiquitous in establishment survey collection and dissemination. As data becomes available at higher frequencies and lower levels of disaggregation, it is prudent to explore modern signal extraction techniques. This work investigates two model-based signal extraction methods with applications to the U.S. Census Bureau’s M3 survey; signal extraction in ARIMA time series (SEATS) and multivariate signal extraction with latent component models. We present both new findings and provide discussion of practical implications. Multivariate methodology allows M3 aggregate series to be viewed jointly by the lower level composition series. For example, we investigate the added benefit to jointly performing signal extraction on automobile manufacturing, light truck/SUV and motor vehicle parts manufacturing to better understand turning points in trend or heteroskedastic seasonal changes of the aggregated motor vehicles and parts series. Here, we utilize component models driven by vector white noise. Additionally, we explore univariate models applying SEATS methodology, a model-based signal extraction paradigm. We compare/contrast seasonally adjusted estimates and components verse current X-11 methodology.