The only thing I remember from courses I had in probability a few years ago is that we also have to clearly defined the event we want to calculate the probability. On the Freakonomics blog, last week, the Israeli lottery was mentioned (here, see also there where I mentioned that, and odds facts from the French lottery), Yesterday, Andrew Gelman claimed (here) that there was a probability error… Well, since Andrew is really a statistician (and a good one… while I am barely an economist), I tried to do the maths…. and to understand where the error was coming from…

Since 6 numbers are drawn out of a pool of numbers from 1 to 37, the total number of combination at each lottery is ```> (n=choose(37,6))
 2324784```

Over 8 lotteries (since there are two draws per week, we can assume there 8 draws per month), the probability of no identical draws is Here is the R code for those who want to check, again,

```> prod(n-0:7)/n^8
 0.999988```

Each month, the probability of “coincidence” (I define “coincidence” the event “over 8 draws, at least two times, we obtained the same 6-uplet” or more precisely (as mentioned here) “over one calendar month, at least two times, we obtained the same 6-uplet“) is

```> (p=1-(prod(n-0:7)/n^8))
 1.204407e-05```

The occurrence of a coincidence each month as a Geometric distribution, with probability p. And it is classical, following Gumbel’s definition (here), to consider 1/p, called the “return period“, i.e. the number of months we have to wait until we observe a coincidence (i.e. a repetition in the same month), since for a geometric distribution ```> 1/p/(12)
 6919.034```

Here, the (expected) return period is 6919 years.

From my point of view, this is “the incident of six numbers repeating themselves within a calendar month”, and this is an event of once in 6919.034 years. On the other hand the median of a geometric distribution is ```> -log(2)/log(1-p)/(12)
 4795.88```

which means that we have 50% chance to get such a coincidence over 4796 years.

Of course, if instead of looking at a longer period, say 100 draws, i.e. one year (here I define “coincidence” the event “over 100 draws, at least two times, we obtained the same 6-uplet“), we have in red the expected return period, and in blue the median of the geometric distribution, ```> M=E=rep(NA,100)
> for(i in 2:100){
+ p=1-exp((sum(log(n-0:(i-1)))-i*log(n)))
+ E[i]=1/p/(100/i)
+ M[i]=-log(2)/log(1-p)/(100/i)
+ }
> plot(1:100,E,ylim=c(0,10000),type="l",col="red",lwd=2)
> lines(1:100,M,col="blue",lwd=2)
> abline(v=8,lty=2)
> points(8,E,pch=19,col="red")
> points(8,M,pch=19,col="blue")```

or below of a log-scaled version As Xi’an did (here), assume now that there is a lottery over 100 countries. Here I define “coincidence” the event “over k lottery draws over 100 around the world, at least two times, we obtained the same 6-uplet“, and then the previous graph becomes (with on the x axis the level of k) Here I have a 12% chance if we consider probability to have identical numbers over a month…

But here, we can have one 6-uplet in Israel, and the other one in Egypt, say… If we want to get the same 6-uplet in the same country, the graph is now i.e. each month there is a chance over one thousand…

```> i=8
> p=1-exp((sum(log(n-0:(i-1)))-i*log(n)))
> 1-(1-p)^100
 0.001203689```

Note: actually, Xi’an mentioned that the probability that this coincidence [of two identical draws over 188 draws] occurred in at least one out of 100 lotteries (there are hundreds of similar lotteries across the World) is 53%! And I got the same,

```> 1-(1-P)^100
 0.5305219```

# Lecture notes on risk and insurance

I just finished some lectures notes on risk and insurance. The notes, that can be downloaded [pdf], are in French, and will be used at the JES (Journées d’Etudes Statistiques), organised at the CIRM (mentioned here). Previous notes on risk measures [pdf] and copulas [pdf]. Again, all comments are welcome… # Names of villages, in France

Keith Briggs published a post here on names of English place name element distribution, which contains almost twenty maps like the one where names ends by -bourn,bourne,burn (here) or -head (there). Actually, it is possible (Robin mentioned that already here) to do similar things in France… Consider the dataset containing the 35,250 commune names (here), it is an xls file containing the official name, the latitude, and on the longitude. To start with something simple, it is possible also to look at village containing “saint” in it  There are a lot, and there is no obvious geographic trend. For some simple geographic trend, t is possible to see where are villages having a name ending with “sur mer” (meaning literally “on the sea”) below on the left. Obviously, we cannot find such places in the Alps. Similarly for names ending with “Seine” they are clearly on the Seine river, on the right  ```> ville=read.table("D:\\r-data\\ville.csv",sep=";",header=TRUE)
> nrow(ville)
 35376
> ville\$maj=as.character(ville\$Nom.Ville)
> n=nchar(ville\$maj)
> I=substr(ville\$maj,pmax(0,n-8),n)
> Ind=I=="-sur-Mer "
> sum(Ind)
 98
> library(maps)
> map('france', fill = FALSE)
> X=ville[Ind,]
> x=as.numeric(as.character(X\$Longitude))
> y=as.numeric(as.character(X\$Latitude))
> points(x,y,pch=19,col="blue",cex=.6)```

In order to continue with some geographic pattern, consider the end of the names, such as “-gny” (below on the left, in red) or “-ac” (below on the right, in blue)  Some pretend that “-ac” comes from Gaelic, and can be found in Celtic regions (here in Brittany). Obviously, there is also an origin in Occitany (south west of France). And this gave also in Oïl region the “-gny” (in North and North-East regions). Consider similarly end of the names, such as “-an” (below on the left, in red) or “-ey” (below on the right, in blue)  Still about the end of the names, it is also possible to look for village ending either with “-a” (below on the left, in red) or “-o” (below on the right, in blue)  We are now in the southern part of France…. “-a” in Corse and Pyrénées, while “-o”can be found in Corse, and in Brittany. For the beginning “ker-” or “lan-”  (below on the left, in red) or “castel-” (below on the right, in blue),  “ker-” appears in 18,000 location names (as mentioned here) but only in some village names. It is similar to “castel-” in the southern part of France.
To go a bit further, 40 years ago, Georges Brassens sang a song entitled “La ballade des gens qui sont nés quelque part“.

He says that people are usually extremely proud of their villages…. Actually, their are more people proud of living over something than under something: below are villages containing “sous” (i.e. under below on the left, in red) or “sur” (i.e. over below on the right, in blue)  On the other hand, villages containing “grand” or “grande” (i.e. tall or big below on the left, in red) or “petit” or “petite” (i.e. small below on the right, in blue) seem to be correlated: close to a city with “grand” there is a village with “petit” in it. For instance Virieu-le-Grand and Virieu-le-Petit, or  Essigny-le-Grand and Essigny-le-Petit.  And finally, I found surprising to see so many village containing “montagne” (ie mountain below on the left, in orange) or starting with “Mont” (below on the right, in purple) that are far from mountains,  You do not need to live close to some mountains to get mountains in it. Even in Brittany you can find dozen of villages starting with “Mont”….