A student who has met the objectives of the course will be able to:
Formulate statistical models (linear, multiple linear), and processes (Gaussian, Poisson and Binomial)
Know the relevant statistical distributions and their uses in the models.
Understand what the statistical parameters of a problem are
Perform inference on the statistical parameters of a problem
Find probabilities and estimates for the statistical parameters
Formulate and evaluate a statistical hypothesis about the parameters
Understand the difference between data and parameters
Perform descriptive analysis and visualization of (current) data.
Perform predictive analysis of (future) data.
Perform regression, with and without estimation of uncertainty
Multiple linear regression
Perform simple statistical analyses using statistical software. We focus on R, but will accept self-taught use of SPSS, MatLab or other relevant software.
Interpret output from software analyses using statistical software packages. We focus on R.
Critically assess results from statistical analyses.
Document and present results in a written report to persons without a statistical background.
Plan, perform and present a simple statistical analysis.
Enable the participants to perform statistical analysis of their own data and to relate the results to practical applications, performing statistical analyses with statistics software. We will follow the theory with practical implementations in the statistical package R, preferably using the students’ own data when possible.
We will start with basic statistical methods and ideas such as condensing the data into statistics and presenting it. Further, we will refresh basic concepts of probability. The statistical concepts and methods will be understood through interpretation of practical application. A mix of bayesian and frequentist concepts and techniques for optimal learning and ability to apply statistical methods with ease and understanding.
Specific content is statistical inference, estimates and hypothesis tests. We will analyse mean and variance from a Gaussian process, proportions, and waiting times from Poisson processes. We will also look at simple and multiple linear regression, and logistics regression.
We will look at test planning, and at interpreting and concluding from the results of a test. There will be a practical project based on the participants' own data consisting of planning, performing and presenting a statistical analysis. This project will form part of the basis for an oral exam.