Unobserved Components Models: Applications in Post-COVID Analysis

Unobserved Components (UCM) is a type of State Space model used to detect and measure changes in a long-term baseline.  This method decomposes time series into components including baseline, linear trend, periodic variations, and irregular.  UCM is often used to analyze variations in baseline values due to change in the state of the system.  In this presentation, UCM is applied to the evolution of the COVID-19 pandemic to determine if various factors such as business conditions have returned to pre-pandemic levels.  Examples include durable goods as a leading indicator, GDP, and unemployment as a lagging indicator.  All major statistical software systems now support UCM; code for this presentation is given in SAS.  The SAS/ETS procedure PROC UCM implements UCM functionality, options, graphical output, and model interpretation. 

About the Presenter

Dr. David J Corliss is a statistical astrophysicist specializing in the dynamics of evolving stellar and cosmic populations. He has worked in the automotive industry for more than 20 years, with extensive work in dynamics of evolving populations of car buyers, reporting and visualization, operations research, big data methods, analytic platform design, and statistical methodology.  Continuing astrophysics research part-time, an important focus of his work has been to bring new developments in academic research to industrial and private sector research.

He presents regularly at local and national SAS events and other conferences, and is active as a leader in the statistics and data science community, writing a monthly column on Data for Good for Amstat News and serving as the non-academic rep for the Statistics section of the American Association for the Advancement of Science. David Corliss is the Founder and Director of Peace-Work, a volunteer cooperative of statisticians and data scientists providing analytic support for charitable groups and applying statistical methods to issue-driven advocacy in Data For Good projects.