`M2 TSi 2013/2014 - Reading Seminar in Statistical Classics - C.P. Robert`

This course is a reading seminar in that all participating students present and discuss one "classic", i.e. one of the important papers that contributed to the advance of Statistics as a science. The evaluation of the participating students is based (a) on the understanding and presentation of the assigned paper, as well as (b) on the contribution to the other discussions. Attendance is thus compulsory for the whole duration of the class. Unless argued otherwise prior to the presentation, the presentation and discussions are in English.

Contacts: Christian Robert, Bureau B638,
tel. 01 4405 4335
email xian@ceremade.dauphine.fr

Papers

The estimation of location and scale parameters of a continuous population of any given form J. Pitman Biometrika (1939)
Periodogram analysis and continous spectra, M.S.Bartlett Biometrika (1950)
Testing for serial correlation in least square regression J. Durbin & G.S. Watson Biometrika (1950)
Monte Carlo sampling methods using Markov chains and their applications, W.K.Hastings Biometrika (1970)
The multiple recapture census for closed populations and incomplete 2^k contingency tables S.E. Fienberg Biometrika (1972)
On the mathematical foundations of theoretical statistics R.A. Fisher Philosophical Trans. Royal Statistical Society London (1922)
On the problem of the most efficient test of statistical hypotheses J. Neyman & E.S. Pearson Philosophical Trans. Royal Statistical Society London (1933)
Algorithm AS 136: A K-Means Clustering Algorithm. J. Hartigan & M. Wong Applied Statistics (1979)
Regression models and life-table D.R. Cox J. Royal Statistical Society (1972)
Bayes Estimates for the Linear Model D.V. Lindley & A.F.M. Smith J. Royal Statistical Society (1972)
Generalized linear models Nelder, J.A. and Wedderburn, R.W. J. Royal Statistical Society (1972)
Marginalisation paradoxes in Bayesian and structural inference A.P. Dawid, M. Stone & J. Zidek J. Royal Statistical Society (1973)
Maximum likelihood from incomplete data via the EM algorithm A.P. Dempster, N.M. Laird and D.B. Rubin J. Royal Statistical Society (1977)
Controlling the false discovery rate: a practical and powerful approach to multiple testing. Benjamini, Y. and Hochberg, Y. J. Royal Statistical Society (1995)
Regression shrinkage and selection via the lasso R. Tibshirani J. Royal Statistical Society (1996)
Bayesian measures of model complexity and fit D.J. Spiegelhalter, N.G. Best, B.P. Carlin, and A. van der Linde J. Royal Statistical Society (2002)
On Rereading R.A. Fisher L. Savage Annals of Statistics (1976)
Bootstrap methods: another look at the jacknife B. Efron Annals of Statistics (1979)
Estimation of the mean of a multivariate normal distribution C. Stein Annals of Statistics (1981)
Estimation of a bounded mean G. Casella & W. Strawderman Annals of Statistics (1981)
Projection pursuit P.J. Huber Annals of Statistics (1985)
Multivariate adaptive regression splines J. Friedman Annals of Statistics (1991)
On the Foundations of Statistical Inference A. Birnbaum J. American Statistical Assoc. (1962)
How biased is the apparent error rate of a prediction rule? B. Efron J. American Statistical Assoc. (1986)
Testing a point null hypothesis: the irreconciability of p-values and evidence J.O. Berger & T. Sellke J. American Statistical Assoc. (1987)
Sampling-based approaches to calculating marginal densities A. Gelfand & A.F.M. Smith J. American Statistical Assoc. (1990)
Adapting to unknown smoothness via wavelet shrinkage. D. Donoho & I. Johnstone J. American Statistical Assoc. (1995)
A decision-theoretic generalization of online learning and an application to boosting Freund, Y. and Schapire, R. J. Computer and System Sciences (1997)
A new look at the statistical model identification H. Akaike IEEE Transactions on Automatic Control (1974)
Support-vector networks C Cortes and V Vapnik Machine learning (1995)

Classes

The class takes place every week on Mondays. In contrast with the other courses in TSi, attendance to all classes is compulsory and absences will impact negatively on the final grade. The selection of the papers will take place during the first class and students are responsible for recovering their allocated paper. As for other courses, cheating and plagiarism are causes for failure of the course and further disciplinary actions.

Reference

Titterington, D.M. and Cox, D.R. (2001) Biometrika: One Hundred Years. Oxford University Press