New search
PFS0176
Multivariate analysis of spectroscopic data for characterization of biomaterials
Repeat of univariate statistics (briefly), repeat of multivariate design principles (briefly), nature of multivariate data, exploratory analysis by principal component analysis and related methods, multivariate classification, multivariate regression, data pre-processing.
After the course, the students will be able to;
- Understand the nature of multivariate data originating from spectroscopic instruments used in research
- Be able to run commercial software, enter data, do calculations and present outcomes as tables and graphs
- Select and apply models for multivariate exploratory data analysis
- Select and apply multivariate classification methods
- Select and apply multivariate regression methods
- Understand the value of tables and graph that can be produced for use in reports, publications, theses
Be able to run and test data pre-processing methods specific to each spectroscopic measurement technique
After the course, the students will be able to;
- Understand the nature of multivariate data originating from spectroscopic instruments used in research
- Be able to run commercial software, enter data, do calculations and present outcomes as tables and graphs
- Select and apply models for multivariate exploratory data analysis
- Select and apply multivariate classification methods
- Select and apply multivariate regression methods
- Understand the value of tables and graph that can be produced for use in reports, publications, theses
Be able to run and test data pre-processing methods specific to each spectroscopic measurement technique
Syllabus and other information
Syllabus
PFS0176 Multivariate analysis of spectroscopic data for characterization of biomaterials, 5.0 Credits
Subjects
TechnologyEducation cycle
Postgraduate levelGrading scale
Pass / Failed
Prior knowledge
MSc or PhD students in forestry or engineering/chemistry/biologyObjectives
To provide a practical course on the use of multivariate data analysis for modelling spectroscopic data. It covers relevant background and theory required to select and analyse an appropriate multivariate method for extracting information about one or more spectroscopic data sets. An ideal starting point for every experimenter who wishes to work effectively, extract maximal information and predict the future behaviour of their system.Content
Repeat of univariate statistics (briefly), repeat of multivariate design principles (briefly), nature of multivariate data, exploratory analysis by principal component analysis and related methods, multivariate classification, multivariate regression, data pre-processing. After the course, the students will be able to; - Understand the nature of multivariate data originating from spectroscopic instruments used in research - Be able to run commercial software, enter data, do calculations and present outcomes as tables and graphs - Select and apply models for multivariate exploratory data analysis - Select and apply multivariate classification methods - Select and apply multivariate regression methods - Understand the value of tables and graph that can be produced for use in reports, publications, theses Be able to run and test data pre-processing methods specific to each spectroscopic measurement techniqueAdditional information
Lectures on the theories behind the methods used. Examples and exercises in the statistical software SIMCA or EVINCE are used to introduce practical aspects and problems often encountered while using multivariate data analysis. One simple group task is performed where students do their own spectroscopic measurement and enter the data to provide a full data analysis report. In addition, individual guidance on multivariate data analysis for a research study is given.Responsible department
Department of Forest Biomaterials and Technology