Process Tracking and Analysis with L_x Exponential Smoothing from Noisy Measurements

Abstract

Tracking process mean from noisy measurements is important in real time data-based analysis since outliers are frequently present in many process measurement systems. This paper discusses robust on-line process mean estimation using L ₁ exponential smoothing (L ₁ -ES). Robustness properties and mean squared error efficiency are investigated. The adaptability of the L _l -ES estimator to processes subject to unknown mean changes is also analyzed and compared with traditional exponentially weighted moving average and median estimators.

Key Words:

• Absolute deviation
• breakdown point
• change-point
• EWMA
• median
• outliers.

Additional information

Notes on contributors

Yi Zhang

Yi Zhang is a Ph.D. candidate at Antai College of Economics and Management, Shanghai Jiao Tong University, China. Her research interests include big data and information quality management.

Xinmin Tian

Xinmin Tian is the Vice-Dean and Associate Professor at Antai College of Economics and Management, Shanghai Jiao Tong University, Executive Vice President of Shanghai Institute of Behavioral Science, and Director of Chinese Leadership Studies Center. He was a recipient of National Distinguished Educator in the Field of Human Resource Management in 2005 and Achievements in Decision-making Policy Recommendations Advisory Research in 2009. His main research lies in strategy human resource management, organizational behavior, applied psychology, leadership, innovation and entrepreneurship.

Wei Jiang

Wei Jiang is the Head and Professor of Operations Management at Antai College of Economics and Management, Shanghai Jiao Tong University, China. Prior to joining Shanghai Jiao Tong University, he worked in AT&T Labs, Stevens Institute of Technology, and Hong Kong University of Science and Technology during the past 15 years. He was a recipient of NSF CAREER award in 2006 and China National Funds for Distinguished Young Scientists award from NSFC in 2013. His main research areas lie in modern quality control and management, data mining and business intelligence, logistics and operations management, information technology and risk management. He is currently serving as Associate Editor for Naval Research Logistics, IIE Transactions, Decision Sciences Journal, etc.