Spectral data and chromatogram data are typical examples of multivariate data, as they measure intensity at multiple wavelengths and time points. By analyzing these data according to multivariate analysis methods, it becomes possible to classify and model the data. In this seminar, we will explain how to perform multivariate analysis on this type of data using JMP products (JMP, JMP Pro) and the benefits of doing so.
*This seminar uses the latest version available at the time of the seminar, "JMP (JMP Pro) 19.1.3". Reports, operation methods, etc. may change with version updates.
(table of contents)
- Introduction and Visualization
- PCA(Principal Component Analysis) and PLS Regression
- Functional principal component analysis using "Functional Data Explorer"
- Analysis example: Relationship between the ratio of ingredients and the shape of the spectrum
1. Introduction and Visualization (14 minutes 25 seconds)
This section explains the characteristics of spectral data in data analysis, how to store the data, and methods for visualization.
2. PCA (Principal Component Analysis) and PLS Regression (15 minutes 55 seconds)
This document explains how to use JMP and generate reports for PCA and PLS regression, which are commonly used methods for analyzing spectral data.
3. Functional Principal Component Analysis using "Functional Data Explorer" (16 minutes 8 seconds)
This article explains functional principal component analysis (PCA) which can capture and classify the "shape of spectral data," which is difficult to achieve with PCA and PLS.
4. Analysis Example: Relationship between the ratio of ingredients and the shape of the spectrum (10 minutes 57 seconds)
As an example of using functional principal component analysis, we will explain how to model the relationship between the spectral composition of a compound and its constituent components, and how to estimate the constituent components from the spectral shape when the constituent components are unknown.
