Statistics and simulation

The course will cover a number of simulation problems and for each of them discuss:

• modeling of the system, often multiple models of the same system.

• implementation of a simulator of the model.

• simulation and data retrieval.

• statistical analysis of simulation results.

• relevance and validity of the results.

Most theoretical themes will be covered through the above practical approach. The following themes will be covered.

Statistics:

• Basic probability calculations: addition rules, multiplication rules, subtraction rule, Bayes' rule. Independent and disjoint events.

• Descriptive statistics: measure of central value and measure of variation.

• Discrete probability distributions: binomial and Poisson distribution.

• Continuous probability distributions: normal distribution and Students's t-distribution. Central limit theorem.

• Interval estimation.

• Hypothesis testing.

• Correlation.

Simulation:

• Use of simulation.

• Modeling of practical problems.

• Relevance and validity of models and simulation results.

• Overview over different simulation methods.

Rationale: Simulation is an important part of development within all disciplines of engineering, and it is a method that requires the involvement of software developers. This module will give the data engineer insight into some common problems from other disciplines, and how computer engineers can help solving them. At the same time, students learn basic and general statistics, a.o. through practical application.

The module:

• covers the curriculum requirement of 5 sp statistics.

• strengthens programming skills through concrete application to practical problems.

• provides some specialist expertise in modeling and simulation.

Learning outcome - knowledge

• knows how statistics may be used in a uniform manner, i.e. how statistics is a necessary tool to measure, describe and evaluate data.

• knows basic probability theory and common probability distributions.

• knows the core theoretical basis for estimation, confidence intervals and hypothesis testing.

• knows different types of simulation and how simulation is applied in engineering disciplines.

Learning outcome - skills:

• can model and simulate basic engineering problems.

• can collect, analyze and present numerical data in general and simulation results in particular.

• can interpret simulation results by means of statistical methods.

• can use statistical principles and concepts.

• has basic skills in developing software for reporting and for graphical presentation.

• masters basic probability theory and can perform estimation, hypothesis testing, simple correlation-/regression analysis.

Learning outcome - general competence

• can obtain relevant answers to technical issues, through the application of statistical studies and methods.

• understands statistical thinking and methods, and can communicate these in written form as well as verbally.

• can communicate with both experts on statistics and users of statistical information on issues within the area.

• understands the potential and limitations of modeling and simulation.

Pedagogiske metoder:

Lectures, exercises and practical assignments. Video lectures for self-study. 

Obligatoriske arbeidskrav:

3 to 6 practical and theoretical assignments/projects that have to be handed in and approved. In some cases, the assignment can be in the form of a presentation in front of the class, and in that case, presence is mandatory for the whole class.

The details of the mandatory assignments will be announced at the beginning of the course.