Seat Projection

Converts vote shares into seats with a highest averages (D'Hondt, Sainte-Laguë) or largest remainder (Hare-Niemeyer) method. With simulations, the uncertainty of the trend is turned into a range of seats.

Usage

popSeats(
  x,
  seats,
  threshold = 0,
  method = c("dhondt", "sainte-lague", "hare"),
  simulations = 0,
  level = 0.9,
  trend = NULL,
  date = NULL
)

Arguments

x

A popPolls object, whose shares are taken from a trend (see popLatest()), or a named numeric vector of vote shares.

seats

Number of seats to distribute.

threshold

Minimum share a party needs to get seats.

method

"dhondt", "sainte-lague" or "hare".

simulations

Number of simulations, 0 for none. Needs a trend with a variance, such as kalman.

level

Coverage of the seat range given by lower and upper.

trend

Name of the trend in x$trends, defaults to the one added last.

date

Date of the projection, defaults to the last date of the trend.

Value

A data.table with the columns party, name (if x is a popPolls object), share and seats, sorted by seats. With simulations also lower and upper, the range of seats, and pSeats, the share of simulations in which the party wins seats.

Details

This is a projection on the national level only: regional constituencies, direct mandates, overhang seats and exceptions from the threshold (such as the basic mandate clauses in Germany and Austria) are not taken into account. For example, seats = 630, threshold = 0.05, method = "sainte-lague" approximates the German Bundestag, seats = 183, threshold = 0.04, method = "dhondt" the Austrian Nationalrat.

The simulations draw the share of every party independently from a normal distribution with the value and variance of the trend. Shares of different parties are in fact negatively correlated, so the ranges are approximate. Use set.seed() for reproducible results.

Examples

popSeats(c(A = 0.35, B = 0.30, C = 0.20, D = 0.10, E = 0.05),
         seats = 100, threshold = 0.06)
#>     party share seats
#>    <char> <num> <int>
#> 1:      A  0.35    37
#> 2:      B  0.30    32
#> 3:      C  0.20    21
#> 4:      D  0.10    10
#> 5:      E  0.05     0

if (FALSE) { # \dontrun{
de = popRead('DE-parliament')
de = popAddTrend(de, name = 'kalman', type = 'kalman')
popSeats(de, seats = 630, threshold = 0.05, method = 'sainte-lague',
         simulations = 2000)
} # }