Time Series Analysis and Spatial Statistics, 7.5 Credits
The course is discontinued
Swedish name: Tidsserieanalys och spatial statistik
This syllabus is valid: 2015-08-24
and until further notice
Course code: 5MS051
Credit points: 7.5
Education level: Second cycle
Main Field of Study and progress level:
Mathematical Statistics: Second cycle, has only first-cycle course/s as entry requirements
Grading scale: Three-grade scale
Responsible department: Department of Mathematics and Mathematical Statistics
Established by: Faculty Board of Science and Technology, 2015-12-16
Contents
The main purpose of the course is that the student should be well aquainted with the basic notions, theory, models and methods for solutions, in time series analysis and spatial statistics. The course covers models for time dependent or spatially dependent data. Such data frequently occurs in financial (e.g. the price development of a merchandise) and scientific (e.g. metheorological observations, radar signales) applications.
The course consists of two parts.
Part 1 (6,5 hp) Theory. The part consists of the general theory of time series, stationary and non-stationary models, e.g. ARMA- and ARIMA-models, prediction of time series, spectral theory, parameter estimation, spectrum and filtration. The part also covers methods for measuring spatial dependence (variogram, covariogram), and techniques for spatial interpolation, especially kriging.
Part 2 (1 hp) Lab Assignments. The part consists of analysis of time series and spatial data using suitable software.
Expected learning outcomes
For a passing grade, the student must be able to
Knowledge and understanding
independently define the notions expectation, covariance function and spectral distribution for time series
independently define parametric ARMA-type models for the mean
thoroughly explain how ARMA-models can be expanded to ARIMA-, FARIMA- and ARCH-models
describe Kalman filtration in general terms
thoroughly explain how variograms and covariograms are derived
thoroughly explain how the (spatial) interpolation methods "The inverse distance method" and "Kriging" work
Skills
identify trends and seasonal variation in time series
derive and estimate expectation, covariance function and spectral distribution for time series, analyse their connections, and derive the uncertainty of the estimates
predict the future of observed time series of different lengths, and critically evaluate the results
apply parametric mean models of ARMA-type, analyse the models' properties and fit real data to them
interpret spatial dependence using variograma and covariograms
predict (spatial) data using the interpolation methods "The inverse distance method" and "Kriging", and critically evaluate the results
Judgement and approach
clearly present the reults from time series analyses and analyses of spatial data, and evaluate the resultats from a scientific perspective
Required Knowledge
The course requires 12 ECTS in Mathematical Statistics and 22,5 ECTS in Calculus including Calculus in several variables and Differential Equations. Proficiency in English equivalent to Swedish upper secondary course English 5/A. Where the language of instruction is Swedish, applicants must prove proficiency in Swedish to the level required for basic eligibility for higher studies.
Form of instruction
The teaching mainly consists of lectures and lessons.
Examination modes
Part 1 is examined by a written exam. Part 2 is examined by written lab reports (U/G). On part 1, one of the following grades is assigned: Fail (U), Pass (G) or Pass with distinction (VG). On part 2, one of the following grades is assigned: Fail (U) or Pass (G). For the whole course, one of the following grades is assigned: Fail (U), Pass (G) or Pass with distinction (VG). In order to receive a passing grade on the course, all parts must be completed with a passing grade. The course grade is decided by the grade on Part 1, and is assigned only when all mandatory examination has been completed. A student who has received a passing grade on a test is not allowed to retake the test in order to receive a higher grade.
A student who has not received a passing grade after participating in two tests has the right to be assigned another examiner, unless there are certain circumstances prohibiting this (see the Higher Education Ordinance, chapter 6, 22§). A request to be assigned another examiner should be adressed to the head of department for the department of mathematics and mathematical statistics. The possibility of being examined based on the current version of the syllabus is guaranteed for at least two years following the student's first participation in the course.
Introduction to Time Series and Forecasting Brockwell Peter J., Davis Richard A. 3rd ed. 2016. : Cham : Springer International Publishing : 2016. : XIV, 425 p. 118 illus., 4 illus. in color. : Table of Contents / Abstracts ISBN: 9783319298542 Mandatory Search the University Library catalogue
Introduction to time series and forecasting Brockwell Peter J., Davis Richard A. 2. ed. (corrected at 8th printing 2010) : New York : Springer : 2002, repr. 2010 : XIV, 434 S. : ISBN: 0-387-95351-5 Mandatory Search the University Library catalogue