**A** definitely brilliant entry on xkcd that reflects upon the infinite regress of producing error evaluations that are based on estimates. A must for the next class when I introduce error bars and confidence intervals!

## Archive for teaching

## I’m getting the point

Posted in Statistics with tags Bayesian statistics, Bayesian textbook, conjugate priors, cross validated, final exam, StackExchange, teaching on February 14, 2019 by xi'an**A** long-winded X validated discussion on the [textbook] mean-variance conjugate posterior for the Normal model left me [mildly] depressed at the point and use of answering questions on this forum. Especially as it came at the same time as a catastrophic outcome for my mathematical statistics exam. Possibly an incentive to quit X validated as one quits smoking, although this is not the first attempt…

## unbiased estimators that do not exist

Posted in Statistics with tags binomial distribution, counterexample, cross validated, exam, existence of unbiased estimators, monomial, teaching, unbiasedness on January 21, 2019 by xi'an**W**hen looking at questions on X validated, I came across this seemingly obvious request for an unbiased estimator of **P**(X=k), when X~**B**(n,p). Except that X is not observed but only Y~**B**(s,p) with s<n. Since **P**(X=k) is a polynomial in p, I was expecting such an unbiased estimator to exist. But it does not, for the reasons that Y only takes s+1 values and that any function of Y, including the MLE of **P**(X=k), has an expectation involving monomials in p of power s at most. It is actually straightforward to establish properly that the unbiased estimator does not exist. But this remains an interesting additional example of the rarity of the existence of unbiased estimators, to be saved until a future mathematical statistics exam!

## p-value graffiti in the lift [jatp]

Posted in Statistics with tags 1960s, basic statistics, course, jatp, lift, p-value, picture, teaching, Université Paris Dauphine on January 3, 2019 by xi'an## a glaringly long explanation

Posted in Statistics with tags ABC, Bayesian Choice, cross validated, exponential families, proof, socks, sufficient statistics, teaching, Thomas Bayes' portrait, undergraduates, Université Paris Dauphine on December 19, 2018 by xi'an**I**t is funny that, when I am teaching the rudiments of Bayesian statistics to my undergraduate students in Paris-Dauphine, including ABC via Rasmus’ socks, specific questions about the book (The Bayesian Choice) start popping up on X validated! Last week was about the proof that ABC is exact when the tolerance is zero. And the summary statistic sufficient.

This week is about conjugate distributions for exponential families (not that there are many others!). Which led me to explain both the validation of the conjugacy and the derivation of the posterior expectation of the mean of the natural sufficient statistic in far more details than in the book itself. Hopefully in a profitable way.

## almost uniform but far from straightforward

Posted in Books, Kids, Statistics with tags counterexample, cross validated, maximum likelihood estimation, order statistics, statistics exam, teaching, The American Statistician, triangular distribution on October 24, 2018 by xi'an**A** question on X validated about a [not exactly trivial] maximum likelihood for a triangular function led me to a fascinating case, as exposed by Olver in 1972 in The American Statistician. When considering an asymmetric triangle distribution on (0,þ), þ being fixed, the MLE for the location of the tip of the triangle is necessarily one of the observations [which was not the case in the original question on X validated ]. And not in an order statistic of rank j that does not stand in the j-th uniform partition of (0,þ). Furthermore there are opportunities for observing several global modes… In the X validated case of the symmetric triangular distribution over (0,θ), with ½θ as tip of the triangle, I could not figure an alternative to the pedestrian solution of looking separately at each of the (n+1) intervals where θ can stand and returning the associated maximum on that interval. Definitely a good (counter-)example about (in)sufficiency for class or exam!

## the decline of the French [maths] empire

Posted in Kids, University life with tags bachelor, Cédric Villani, French education ministry, Le Monde, measure theory, programmes, secondary schools, teaching on December 16, 2017 by xi'an**I**n Le Monde edition of Nov 5, an article on the difficulty of maths departments to attract students, especially in master programs and in the training of secondary school maths teachers (Agrégation & CAPES), where the number of candidates usually does not reach the number of potential positions… And also on the deep changes in the training of secondary school pupils, who over the past five years have lost a considerable amount of maths bases and hence are found missing when entering the university level. (Or, put otherwise, have a lower level in maths that implies a strong modification of our own programs and possibly the addition of an extra year or at least semester to the bachelor degree…) For instance, a few weeks ago, I realised for instance that my third year class had little idea of a conditional density and teaching measure theory at this level becomes more and more of a challenge!