By a news market is meant an electronic market where bets can be taken on various future events. Several examinations have shown that such markets can gather dispersed partial information from their actors into a single rate of exchange (information aggregation) and thereby predict events. The study sets out to present a model of these news markets and their actors and examine the information aggregation. The model devised describes the thinking and behaviour of the market actors and runs simulations in which agents based on the models trade. The findings confirm the model's validity and draw attention to some interesting facets of how information aggregation works.