Statistics - Statistics Research

Here are some of the ongoing research projects in statistics at IBM.

  • spatial modeling [details to be added]

  • L1-norm spectral analysis. A new type of periodogram, called the Laplace periodogram, can be derived by replacing least squares with least absolute deviations in the harmonic regression procedure that produces the ordinary periodogram of a time series. The Laplace periodogram can be used as a nonparametric tool for analyzing the serial dependence of time-series data. Superiority of the Laplace periodogram in handling heavy-tailed noise and nonlinear distortion has been demonstrated by simulation. A real-data example shows its great effectiveness for analyzing heart rate variability in the presence of ectopic events and artifacts. The approach is described in a paper available on Ta-Hsin Li's personal research page.

  • L-moments. Research continues in the theory and practice of L-moments -- summary statistics for probability distributions and data samples, computed from linear combinations of the ordered data values. For more details see the L-moments page.