Yesterday, I was asked how to write a code to generate a compound Poisson variables, i.e. a series of random variables  where  is a counting random variable (here Poisson disributed) and where the 's are i.i.d (and ind... [Read more...]

Quantifying the success of government policies is clearly important. Randomized control trials, like those conducted by drug companies, are often described as the ‘gold-standard’ for policy evaluation. Under these, a policy is implemented in/to one area/group (treatment), but not in/to another (control). The difference in outcomes between ... [Read more...]

On http://www.bakadesuyo.com, there was recently an interesting discussion about infidelity, the key question being "at what ages are men and women most likely to have affairs?" The discussion is based on some graphs, e.g. The source is a paper b... [Read more...]

I am always surprised to see many people on Twitter tweeting about #opendata, e.g. @data4all, @usdatagov, @datapublicatwit, @ProPublica or @open3 among so many others... Initially, I was also very enthousiastic, but I have to admit that open data a... [Read more...]

Yes, I did just mix English, Spanish and French. And no, I living the “fishy” life, popular opinion to the contrary. Here’s the story. As someone who spends the majority of his time working online, with no oversight, I notice that I tend to drift a lot. I don’... [Read more...]

Coded as follows: x = seq(.001,50,.001) par(bg="black") par(mar=c(0,0,0,0)) plot(x,sin(1/x)*rpois(length(x),x),pch=20,col="blue") [Read more...]

Maybe you remember playing this one as a kid. If you are about my age, you may have even created a version of this game as one of your first computer programs. You guess a number, the computer tells you if you if you are too low or high. I’... [Read more...]

Click on image for a larger version. Here is the code: par(bg="black") par(mar=c(0,0,0,0)) plot(sort(rpois(10000,100))/rpois(10000,100),frame.plot=F,pch=20,col="blue") [Read more...]

A common model for counts data is the Poisson. There are cases however that we only record positive counts, ie there is a truncation of 0. This is the truncated Poisson model. To study this model we only need the total counts and the sample size. This comes from the sufficient ... [Read more...]