Skip to main content
SciX 2025 Main banner

Oral Contributed Presentation

CHEM

CHEM05: Recent Advances in Chemometrics

How to Decode Rainbows: The Best and Least Confusing Ways to Analyze Spectral Data

Wednesday, October 8, 2025
2:10 PM - 2:30 PM EDT
Location: Meeting Room #7

    Presenting Author(s)

  • WW

    William J. Worley

    Principal Systems Engineer
    JMP Statistical Discovery, LLC
    Mason, Ohio, United States

    • Project Lead(s)

  • PH

    Peter J. Hersh, PhD

    Senior Systems Engineer
    JMP Statistical Discovery, LLC
    Denver, Colorado, United States

    • Non-Presenting Author(s)

  • TS

    Thomas Skov

    Principal Scientist
    Novonesis
    Hoersholm, Hovedstaden, Denmark

In fifteen words or less, explain the significance of this contribution (Novel Aspect).: Functional Data Analysis is making a splash for analyzing chemometric data!

Abstract Text:

Analyzing spectral data is a bit like trying to decode a rainbow—it’s beautiful but full of tricky surprises.  Spectral data presents unique challenges due to its highly correlated nature, which renders many conventional techniques ineffective. In this talk, we will identify these challenges and explore advanced methods tailored for handling such data. Specifically, we’ll dive into three powerful techniques: Principal Component Analysis (PCA), Partial Least Squares (PLS), and Functional Data Analysis (FDA). By comparing these methods, we’ll highlight their strengths, limitations, and practical applications, offering insights into choosing the best approach for analyzing highly correlated spectral data.  This will ensure you can transform your data into a vibrant spectrum of success even if you never look at a rainbow the same way.