Advanced Statistics for Analytical Chemistry

6 EC

Semester 2, period 5

5254ASFA6Y

Owner Master Chemistry (joint degree)
Coordinator dr. J.A. Westerhuis
Part of Master Chemistry (joint degree), track Analytical Sciences, year 1

Course manual 2017/2018

Course content

The size of the laboratory data is nowadays booming: instruments produce a large amount of data. The job market in analytical chemistry is demanding professionals with a deep knowledge on how to treat such data, in order to extract the maximum information.

The course contains three parts. The first part is dedicated to Bayesian statistics, its theory and applications. The second is dedicated to signal processing methods in chemistry, including smoothing, alignment, feature detection, scatter correction, etc.. The third part is dedicated to advanced multivariate calibration and classification methods. These include LDA, QDA, k-NN, Naive Bayes, SVM, as well as PLS-DA, n-way PLS, and GLM models.

Study materials

Literature

  • D.L. Massart et al. “Handbook of chemometrics and qualimetrics”, parts A and B. Elsevier.
  • T. Naes et al. “Multivariate calibration and classification”, NIR publications.
  • C.E. McCulloch, S.R. Searle, J.M. Neuhaus, “Gneralized, Linear, and mixed models, 2nd Ed.”, Wiley.
  • T. Hastie, R. Tibshirani, J. Friedman, “The elements of statistical learning, 2nd Ed.”, Springer.
  • A. Smile, R. Bro, P. Geladi, “Multi-way analysis: applications in the chemical sciences”, Wiley.
  • R.G. Brereton, “Chemometrics: Data analysis for the laboratory and chemical plant”, Wiley.
  • D.S. Sivia, J. Skilling, “Data analysis: A Bayesian tutorial, 2nd Ed.”, Oxford science publications.
  • N. Armstrong, D.B. Hibbert, “An introduction to Bayesian methods for analyzing chemistry data. Part :An introduction to Bayesian theory and methods”, Chemom. Intell. Lab. Syst. 97 (2009) 194–210.
  • D.B. Hibbert, N. Armstrong, “An introduction to Bayesian methods for analyzing chemistry data. Part II: A review of applications of Bayesian methods in chemistry”, Chemom. Intell. Lab. Syst. 97 (2009) 211–220.
  • S.D. Brown, R. Tauler, B. Walczak, “Comprehensive Chemometrics”, Elsevier.

Other

  • Presentations slides.

Objectives

Students have insight into the main methods used for data treatment in industry.

Student understands complexity of multivariate analytical chemical data.
Students know how standard multivariate methods as PCA and PLS work.
Students can interpret multivariate models of multivariate data.
Students know why validation is necessary and how it is applied.
Students are able to apply the data analysis methods to new data.
Students are able to critically assess scientific papers in which these methods have been applied.
Students are able to present their data analysis results in a scientifically sound manner.

Teaching methods

  • Lecture
  • Seminar
  • Computer lab session/practical training

1. Theoretical lectures + discussion.
2. Werkcolleges with data sets to be analysed. This includes an advanced Matlab programming, building your own functions in Matlab to sovle complex problems.
3. An assigment, including a complex & challenging real data set should be solved in paralell. The data set will consists of data set problems from industry that should be solved using the material explained in class, building up your own Matlab functions and investigating other techniques (maybe not discussed in class, but related to the material discussed) in class. The assignment will be covering one (or several) of the following aspects: Signal processing, Bayesian, Multivariate statistics.
4. Brain-storming sessions will be run periodically to solve the assignment in which all students will participate.

Learning activities

Activity

Number of hours

Computerpracticum

30

Hoorcollege

30

Presentatie

4

Tentamen

3

Zelfstudie

101

Attendance

The programme does not have requirements concerning attendance (OER-B).

Assessment

Item and weight Details

Final grade

1 (100%)

Tentamen

Fraud and plagiarism

The 'Regulations governing fraud and plagiarism for UvA students' applies to this course. This will be monitored carefully. Upon suspicion of fraud or plagiarism the Examinations Board of the programme will be informed. For the 'Regulations governing fraud and plagiarism for UvA students' see: www.student.uva.nl

Timetable

The schedule for this course is published on DataNose.

Additional information

Prior knowledge: Fundamental Analytical Sciences. This course is an extension of the course 'Fundamental of Analytical Sciences' and it constitutes its second part.

Contact information

Coordinator

  • dr. J.A. Westerhuis