Estatística Numérica Computacional
Objectives
Students should be able to understand and apply the following statistical methods which need intensive use of the computer: algorithms of type Newton-Raphson, Monte Carlo, resampling techniques (Bootstrap e Jackknife), sampling-resampling techniques and iterative simulation (Monte Carlo via Markov Chain, MCMC method). Students should be able to use the statistic software R in applied problems, by using the adequate available libraries or by adequately modifying them in case of necessity. Students should further be able to write reports where, using statistical techniques, a full analysis of the study case is done and well justified conclusions are drawn.
General characterization
Code
12023
Credits
4.0
Responsible teacher
Isabel Cristina Maciel Natário
Hours
Weekly - 3
Total - 41
Teaching language
Português
Prerequisites
Basic notions of Analysis and intermediate level notions of Probability and Statistics.
Bibliography
Albert J., Bayesian Computation with R, Springer, 2007.
Crawley, M. J., The R Book, 2nd Edition, Wiley, 2012.
Davison, A.C., Hinkley, D.V., Bootstrap Methods and their Application, Cambridge University Press, 1997.
Gamerman, D., Lopes, H.F., Stochastic Simulation for Bayesian Inference, Chapman & Hall/CRC, 2006.
Gentle, J.E., Random Number Generation and Monte Carlo Methods, Springer-Verlag, 1998.
Gentle, J.E., Computacional Statistics, Springer, 2009.
Hossack, I.B., Pollard, J.H., Zehnwirth, B., Introductory Statistics with Applications in General Insurance, Cambridge University Press, 2nd Edition, 1999.
Rizzo, M.L., Statistical Computing with R. Chapman & Hall/CRC, 2007.
Robert, C., Casella, G., Introducing Monte Carlo Methods with R, Springer, 2010.
Ross, S.M., Simulation, 3rd Edition, Academic Press, 2002.
Teaching method
The curricular unit is divided into 4 learning units that include each 2-3 videos of oral exposition of the contents, of 10-15 minutes each, along with the presentation of examples and complemented by solved proposed exercises. In the end of each learning unit a revising exercise is delivered. During the learning units presentation two evaluation assignments are distributed, contributing for the final grading. A timetable for explaining doubts to students made is available via Skype.
Evaluation method
Two theoretical-practical assignments, needing R programming, one for each 2 curricular units (50% + 50%).
Subject matter
1. Pseudo-random number generation (discrete and continuous).
2. Newton-Raphson method.
3. Variance reduction techniques.
4. Resampling techniques: Bootstrap and Jackknife.
5. Monte Carlo methods.
6. Numerical optimization for the maximum likelihood method.
7. Fisher scoring method.
8. Sampling-resampling methods.
9. Monte Carlo via Markov Chain (MCMC) methods: the Gibbs Sampler and Metropolis Hastings algorithms.
10. Use of the taught techniques and adaptation of the libraries to practical case studies.
11. Writing of reports where, using statistical techniques, a full analysis of some case studies is done and conclusions drawn.