Latent-variable regression for spectroscopic measurements.
Event details
| Date | 25.09.2009 |
| Hour | 10:15 |
| Speaker | M. P. Gujral, Laboratoire d'Automatique, EPFL |
| Location |
MEC2405
|
| Category | Conferences - Seminars |
Latent-variable regression has found use in many diverse applications such as social science, business studies and engineering. I will focus the attention on the engineering application in spectroscopy (e.g. near-infrared, infrared, Raman, chromatography, nuclear magnetic resonance, mass spectroscopy), where a typical goal is to estimate analyte concentrations from the corresponding spectra. Spectroscopic data is characterized by (i) a high degree of multicollinearity in the spectra, and (ii) a linear relationship between the spectra and the concentrations. These properties make latent-variable regression models, such as PCR and PLSR, ideal candidates for calibration. The talk will be tutorial in nature. In the first part, I will introduce PCR and PLSR, the statistics (or heuristics) behind these methods, and a cautionary note on some pitfalls or common misuses of these methods. In the second part, I will present my thesis work on the correction or update of PCR/PLSR models when affected by systematic disturbances.
Links
Practical information
- General public
- Free
Contact
- Gregory Francois