S. Merrill, A PROBABILISTIC MODEL FOR THE SPATIAL-DISTRIBUTION OF PARTY SUPPORT IN MULTIPARTY ELECTORATES, Journal of the American Statistical Association, 89(428), 1994, pp. 1190-1197
Spatial models of electoral competition locate voters and parties at p
oints in euclidean space-representing issue positions-and specify util
ity of voters for parties as functions of these positions. Utility fun
ctions may also have stochastic components unassociated with issues. I
n this article probabilistic models are compared in which the utility
function incorporates distance between voter and party positions (prox
imity model) or a scalar product (directional model). Model specificat
ion is significant because of its relation to party strategy and the r
esulting spatial distribution of parties. Maximum likelihood is used t
o estimate parameters of a mixed directional and proximity model-with
stochastic and strategic components-from data in Norwegian and Swedish
election studies. Expected spatial distributions of voters by party s
upport are determined for the multiparty electorates of Norway and Swe
den. Unlike previous deterministic work, which strongly favors the dir
ectional model, the results obtained here suggest that a mixture of pr
oximity and directional probabilistic models may provide a substantial
ly better fit than either pure model or a deterministic model.