Statistical Modelling and Inference 10 ECTS-credits Learning Outcomes On completion of this course, students will be able to: - analyze data with the R software using models for non-normal outcomes, repeated measurement, survival analysis and missing data - analyze data with the R software using Bayesian sampling methods - critically evaluate the output from these analyses The students will be able to demonstrate knowledge of: - statistical inference for a broad range of statistical models - the assumptions behind the models - the philosophical differences between statistical and machine-learning methods - the philosophical differences between schools in statistics: frequentist, likelihoodist and Bayesian Course Content Elements of likelihood inference, aspects of frequentist properties, profile likelihood, evidence and the likelihood principle, Bayesian methods, extended likelihood principle, model selection, dealing with nuisance parameters, survival analysis, classification of missing data mechanisms, dealing with missing data using the EM algorithm and multiple imputation, mixed linear models. Assessment Written examination and home exercises. Forms of Study Lectures and computer exercises. Prerequisites Probability Theory and Markov Processes Statistical Computing with R Linear and Generalized Linear Models Literature Pawitan, Y.. (2001) In all likelihood : statistical modelling and inference using likelihood. Oxford : Clarendon. (528 pp). Gelman A., Carlin J.B., Stern H.S. and Rubin, D.B. (2004) Bayesian Data Analysis. Second Edition. Chapman & Hall (689 pp). Reference Literature Hastie, T., Tibshirani, R. and Friedman, J. (2009) The elements of statistical learning: data mining, inference and prediction. Second Edition. Springer. Efron, B. 1987. Better bootstrap confidence intervals (with discussion). Journal of the American Statistical Association 82, 171-200. Efron, B. 1998. R.A. Fisher in the 21st century (with discussion). Statistical Science, 13, 95-122. Schweder, T. 2003. Abundance estimation from photo-identification data: confidence distributions and reduced likelihood for bowhead whales off Alaska. Biometrics, 59, 976-985. Schweder, T. and Hjort, N.L. 2002. Confidence and likelihood. Scandinavian Journal of Statistics, 29, 309-332. Bjornstad, J., Predictive likelihood: A Review (with discussion), Statistical Science vol 5, 1990, 242- 265 Bjornstad, J., On the generalization of the likelihood function and the likelihood principle, JASA vol 91, 791 – 806 Ytterstad, E. (1991). Predictive likelihood predictors for the AR(1) model: a small sample simulation, Scand. J. Statist vol 18, 97-110