NFYK13011U Applied Statistics: From Data to Results
MSc Programme in Physics
MSc Programme in Nanoscience
MSc Programme in Environmental Science
MSc Programme in Physics w. minor subject
The course will give the student an introduction to and a basic
knowledge of statistics and data analysis. The focus will be on the
application of statistics and thus proofs are omitted, while
examples and use of computers take their place. For this reason,
programming plays a central role and is an essential requirement
The course will cover the following subjects:
- Introduction to statistics.
- Distributions - Probability Density Functions.
- Error propagation.
- Monte Carlo - using simulation.
- Statistical tests.
- Parameter estimation - philosophy and methods of fitting data.
- Chi-Square and Maximum Likelihood fits.
- Simulation and planning of an experiment.
- Multidimensional data and Fisher Discriminant.
- Introduction to Machine Learning.
- The power and limit of statistics.
The student should in the course obtain the following skills:
- Determining mean, width, uncertainty on mean and correlations.
- Understanding how to use probability distribution functions.
- Be able to calculate, propagate and interpret uncertainties.
- Be capable of fitting data sets and obtain parameter values with uncertainties.
- Know the use of simulation in planning experiments and data analysis.
- Select and conduct appropriate statistical tests.
The student will obtain knowledge about statistical concepts and procedures, more specifically:
- Binomial, Poisson and Gaussian distributions and origins.
- Error propagation formula – use and applicability.
- ChiSquare as a measure of Goodness-of-fit.
- Calculation and interpretation of p-values.
- Determination and treatment of potential outliers in data.
- The applicability of Machine Learning.
This course will provide the students with an understanding of statistical methods and knowledge of data analysis, which enables them to analyse data from essentially ALL fields of science. The students should be capable of handling uncertainties, fitting data, applying hypothesis tests and extracting conclusions from data, and thus produce statistically sound scientific work.
See Absalon for final course material. The following is an example of expected course literature.
Primary literature: Statistics - A Guide to the Use of
Statistical Methods in the Physical Sciences, Roger Barlow.
Additional literatur: Statistical Data Analysis, Glen Cowan. Data Reduction and Error Analysis, Philip R. Bevington. Statistical Methods in Experimental Physics.
Academic qualifications equivalent to a BSc degree is recommended.
There will be an introduction the week before the course begins. You will be informed about time and place later (on the course webpage and by Email).
- Theory exercises
- 7,5 ECTS
- Type of assessment
- Continuous assessmentWritten assignment, 28 hoursThe final grade is normally given based on the continuous evaluation as well as on the take-home exam with the following weight;
20% from a project, 20% from a problem set, and 60% from a 28 hours take-home exam.
It is possible to some extent to arrange a different weight in individual cases in agreement between the student and course responsible, if this can be justified.
- All aids allowed
- Marking scale
- 7-point grading scale
- Censorship form
- No external censorship
The exam form is identical to the regular exam; the project and/or problem set that were approved during the course can be re-used. The remaining project and/or problem set should be approved 2 weeks before the re-exam.
Criteria for exam assesment