This afternoon, we’ve seen in the training on data science that it was possible to use AIC criteria for model selection. __ library(splines) __ AIC(glm(dist ~ speed, data=train_cars, family=poisson(link="log"))) [1] 438.6314 __ AIC(glm(dist ~ speed, data=train_cars, family=poisson(link="identity"))) [1] 436.3997 __ AIC(glm(dist ~ bs(...

A few years ago, I did publish a post to visualize and empirical claims frequency in a portfolio. I wanted to update the code. Here is a code to get a dataset, sinistre

In order to illustrate the problem of chosing a classification model consider some simulated data, __ n = 500 __ set.seed(1) __ X = rnorm(n) __ ma = 10-(X+1.5)^2*2 __ mb = -10+(X-1.5)^2*2 __ M = cbind(ma,mb) __ set.seed(1) __ Z = sample(1:2,size=n,replace=TRUE) __ Y = ma*(Z==1)+mb*(Z==2)+rnorm(n)*5 __ df = data.frame(Z=...

Classification trees are nice. They provide an interesting alternative to a logistic regression.  I started to include them in my courses maybe 7 or 8 years ago. The question is nice (how to get an optimal partition), the algorithmic procedure is nice (the trick of splitting according to one variable, and only ...

Yesterday evening, I discovered some interesting slides on False-Positives, p-Hacking, Statistical Power, and Evidential Value, via @UCBITSS ‘s post on Twitter. More precisely, there was this slide on how cheating (because that’s basically what it is) to get a ‘good’ model (by targeting the p-value) As mentioned by @david_...

Disclamer: This is a joint post with Avner Bar-Hen, a.k.a. @a_bh, Benjamin Guedj, a.k.a. @bguedj and Nathalie Villa, a.k.a. @Natty_V2 Organised annually since 1970 by the French Society of Statistics (SFdS), the Journées de Statistique (JdS) are the most important scientific event ...

This Tuesday, I will be in Leuven (in Belgium) at the ACP meeting to give  a talk on Data Science: from Small to Big Data. The talk will take place in the Faculty Club from 6 till 8 pm. Slides could be found online (with animated pictures). As usual, comments are welcome.