Dr. Ta-Hsin Li is a Research Staff Member. He received the Ph.D. degree in applied mathematics from the University of Maryland, College Park, in 1992. Before joining IBM in 1999, he was on the faculty of the Statistics Department at Texas A&M University, College Station (1992–1997) and the Statistics and Applied Probability Department at the University of California, Santa Barbara (1998–2000). His main research interests are time series analysis, statistical signal processing, statistical and AI machine learning methods for business applications. He serves as Associate Editor for the IEEE Transactions on Signal Processing (2000–2006, 2009–2012), the EURASIP Journal on Advances in Signal Processing (2006–present), the Journal of Statistical Theory and Practice (2011–present), Technometrics (2013–2016), and Applied Stochastic Models for Business and Industry (2016-present).
Dr. Li is Fellow of the American Statistical Association (ASA).
Current Research Areas:
- Statistical, AI, and machine learning methods for business applications
- Quantile regression methods for spectral analysis of time series
- Statistical theory and methods for time series analysis
- Statistical methods for forecasting
Quantile-Frequency Analysis (QFA):
Quantile-frequency analysis (QFA) is a nonlinear spectral analysis method for time series. It is based on quantile periodograms constructed from trigonometric quantile regression. Computer code is available here together with some experimetal data.
- Quantile-frequency analysis and deep learning for signal classification.
- Quantile-frequency analysis and spectral measures for diagnostic checks of time series with nonlinear dynamics.
- Quantile Fourier transform, quantile series, and nonparametric estimation of quantile spectra.
- From zero crossings to quantile-frequency analysis of time series with an application to nondestructive evaluation.