Modelling Information Incorporation in Markets, with Application to Detecting and Explaining Events

We develop a model of how information flows into a market, and derive algorithms for au­ tomatically detecting and explaining relevant events. We analyze data from twenty-two "political stock markets" (i.e., betting mar­ kets on political outcomes) on the Iowa Elec­ tronic Market (IEM). We prove that, under certain efficiency assumptions, prices in such betting markets will on average approach the correct outcomes over time, and show that IEM data conforms closely to the theory. We present a simple model of a betting market where information is revealed over time, and show a qualitative correspondence between the model and real market data. We also present an algorithm for automatically de­ tecting significant events and generating se­ mantic explanations of their origin. The al­ gorithm operates by discovering significant changes in vocabulary on online news sources (using expected entropy loss) that align with major price spikes in related betting markets.