From Real Time Economics:
Beth Ann Bovino came out tops in The Wall Street Journal’s ranking of forecasters, but she was already a member of a rarified group: Women economists working on Wall Street.
Indeed, of the 51 forecasts included in the 2013 survey, only six
came from women. But a quick perusal of the women’s forecasts shows that
they , on balance, did a pretty good job of figuring out what the year.
On average, they scored 87.9 on the Journal’s ranking methodology,
which compared to an average score of 81. (See the full rankings.)
But what are the chances that six randomly selected economists would
score as well as the women did? Since there are nearly 13 billion ways
to draw six economists out of 51, getting an exact number would be
difficult. But running a Monte Carlo simulation — running a program that
draws 6 forecasts at random from the 51 over and over again — is a way
to come up with a reasonably precise estimate.
For the first simulation, we had the computer draw 10,000 sets of six
forecasts randomly and check to see if the average off those six
forecasts outscored the women’s average. And we had the computer repeat
this process 1,000 times (for a total of 10 million draws.)
The following histogram shows that the women got outscored between
18% and 20% of the time, with a concentration around 19%. Not bad.
But maybe not entirely fair — to the women — either. After all, by
drawing from the entire set of forecasts, sometimes we were drawing one
or more of the women’s forecasts, which would tend to flatter the
average score....MORE
