02423 Advanced Dataanalysis and Statistical Modelling |
Danish title: Videregående dataanalyse og statistisk modellering |
Language: Danish ECTS-creditpoints: 10, External examination.
|
|
Exam schedule:
|
F2-A (maj 29 2002) |
Recommended semester: 4th -7th semester |
Scope and form: Lectures, student team work, self-study of selected topics, project work. |
Evaluation: Written exam and approval of reports
|
Examination: 13-scale |
Previous course: 04341 |
Prerequisites: Knowledge of elementary statistical methods (course 02401 / 04040 / 04041) and acquaintance with the so-called general linear model, GLM, (course 02409 / 04241). Furthermore elementary knowledge of the SAS software system is assumed. |
Aim: To introduce the participants to concepts and methods that are used for analysis of systematically recorded data, and to give competence to use these methods in practice. By the end of the course you should be able to analyze variation in data, identify significant sources of variation and assess their effect, usually by means of a statistical model. You should be able to fit non-linear models to data in relation to the problem under investigation and under due consideration to the characteristics of the data, to the uncertainty associated with individual datapoints and to possible hierarchical structures induced by the data generation. Moreover, you should be able to interpret and use various model fitting methods, in particular the Bayes-pardigm for incorporation of prior knowledge. |
Contents: Survey sampling, sample selection. Model fitting, likelihoodfunction and the Bayesian paradigm, statistical decision theory. Model validation, resampling methods. Distributions of response: binomial-, Poisson-, multinomial-, exponential-, gammadistributions, and inverse Gaussian distribution. Generalised linear models, hierarchical (mixed) generalised linear models, empirical Bayes methods, BLUP estimation. Models for lifetime- and survival data, censored observations, models with proportional hazards. |
Contact: Poul Thyregod, building 321, (+45) 4525 3361, pt@imm.dtu.dk |
Department: 002 Informatics and Mathematical Modelling |
Course URL: http://www.imm.dtu.dk/courses/02423 |
Keywords: Survey sampling, generalized linear models, life-time data, hierarchical variation, Bayes paradigm |
Updated: 20-04-2001 |
|
|