I am not a big fan of those ceremonies, where some actors pretend that they are extremely happy to be there, and then some win a trophy, some don’t, and those who win start to cry, and those who did not get a trophy try to pretend that they are not affected, etc. The other reason is that, since I have several kids, I do not go to see the movies that often (I mean apart from Shrek, Toy Story… Harry Potter is probably the only movie I’ve seen with real actors – or at least human actors).
But I remember being surprised when I looked at the nominees in newspapers,
Actresses are beautiful and look young, while actors are more experienced. So I have try to see how old were those who win an Oscar, as best actor (here) or best supporting actor (there), and best actress (here) and best supporting actress (there).
OSCAR=read.table("http://freakonometrics.blog.free.fr/public/data/OSCAR.csv", sep=",",header=TRUE,dec=".") actor=OSCAR[,1] suppactor=OSCAR[,2] actress=OSCAR[,3] suppactress=OSCAR[,4] actor=actor[is.na(actor)==FALSE] actor=actor[actor>0] actress=actress[is.na(actress)==FALSE] actress=actress[actress>0] suppactor=suppactor[is.na(suppactor)==FALSE] suppactor=suppactor[suppactor>0] suppactress=suppactress[is.na(suppactress)==FALSE] suppactress=suppactress[suppactress>0] boxplot(actor,suppactor,actress,suppactress,col=c("blue","blue","red","red"), names=c("actor","supp. actor","actress","supp. actress"))
On average, a best actress is 36 years old, while a best actor is 44 years old. Which is quite a difference… Perhaps because it takes more time to an actor to be a good one ? Assuming that they start acting at 18, it takes 18 more years for an actress to be recognized as a good one (here the best one), and 26 for an actor. Or perhaps it is simply because leading actresses have to look young…
The oldest actor who won an Oscar was Henry Fonda (at the age of 76) and the oldest actress was Jessica Tendy (nearing 81). Tatum O’Neal became the youngest person to win the best suppo
rting actress award at the age of 10 (she was 8 when she was acting). The youngest best actress was Marlee Matlin, 21. The distribution was be seen below, with actors in blue, and actresses in red, best supporting actors in dotted lines, and best actors in plain lines,
plot(density(actor),xlim=c(10,80),axes=FALSE, col="blue",names="",ylab="",xlab="",ylim=c(0,.051)) lines(density(suppactor),col="blue",lty=2) lines(density(actress),col="red") lines(density(suppactress),col="red",lty=2) axis(1)
Note that the age of supporting actors is older that leading ones. E.g. the average age for supporting actors winning an Oscar is 50, while it is 44 for actors. Similarly, it is 40 for supporting actresses, and 36 for actresses.
> mean(suppactor) [1] 50.23762 > mean(actor) [1] 44.29982 > mean(suppactress) [1] 40.55766 > mean(actress) [1] 36.39733
Here, I have to admit that I was surprised. I always thought that being a supporting actor was a first step before being a leading one. So winners of supporting awards should have been younger that winners of leading ones. But this is not the case.
And the dynamic here is rather stable, with actors,
and actresses,
except that the age difference between supporting roles and leading roles have increased in the 80’s for actors, while it decreased in the 80’s for actresses.
, the year (of the marriage), and on column 
, the age of the man when he gets married. Assume that those were rawdata, i.e. that we have the number of marriages of men of age 
, we want to estimate (or predict) the age he will get married, if he gets married. With raw data, we can do it… The first step is to build up triangles (to have a cohort vs. age lecture of the data), and then to consider a model, e.g.
is a year effect, and 
is a cohort effect.
and 
, where now 
denotes the cohort.
is the following
, but unfortunately, I do not think any interpretation is valid (unless demography did not change last century). For instance, the following sum
