Danish general election prediction competition
This is for submitting entries for predictoin the distribution of legal votes in the upcoming Danish election. The rules of the competition is as follows:
-
The competition is about predicting the votes given to each of the parties eligible for the Danish parliament.
-
The deadline for submitting entries is May 31st 2019 at midnight (CET).
-
A submission consists of a pull request to this repository with an R data object saved with
save()
. Name theR
object and the file the same, and don't use a name that is already in use. -
Each person should only submit one entry
-
A prediction consists of a data frame with two columns:
party
andpercent
. Thepercent
column will subsequently be scaled such that it sums to 100.A blueprint for an
R
object to submit ismypred <- data.frame(party=c("A", "B", "C", "D", "E", "F", "I", "K", "O", "P", "V", "OE", "AA"), percent=rep(1/13, 13))
-
Each prediction is score by the Brier score: the mean squared prediction error when comparing to the actual distribution of votes after the election. Smallest score wins.