O'Reilly logo

The Handbook of News Analytics in Finance by Leela Mitra, Gautam Mitra

Stay ahead with the world's most comprehensive technology and business learning platform.

With Safari, you learn the way you learn best. Get unlimited access to videos, live online training, learning paths, books, tutorials, and more.

Start Free Trial

No credit card required

12.2 BACKGROUND LITERATURE

In his seminal work, Clark (1973) proposed that a mixture of normal distributions should be utilized to model the empirical distribution of security price changes. Clark's model assumes that news events are important for pricing securities and that news arrives at a random rate over the trading period. This is normally referred to in the literature as the Mixture of Distribution Hypothesis (MDH). Using the same assumptions, Tauchen and Pitt (1983) and Harris (1986, 1987) show that the joint distribution of trading volume and price changes can be modeled by a mixture of bivariate normal distributions. More specifically, in the standard MDH model, the daily price change and trading volume are the sum of independent intraday price changes and volume that occur as a result of the arrival of information events. After each event, price changes as a result of traders responding to the new information. Intraday price changes and the volume of those trades are generally assumed to be jointly independent and identically distributed with finite variance. If the number of news event arrivals are sufficiently large for a given interval then, following the Central Limit Theorem, the joint distribution of price changes and trading volume are approximately bivariate-normal and conditional on the number of information events. Thus, the conditional variance of price changes is an increasing function of the rate of information flow on the market.

Building on the MDH, Lamoureux ...

With Safari, you learn the way you learn best. Get unlimited access to videos, live online training, learning paths, books, interactive tutorials, and more.

Start Free Trial

No credit card required