© 1999 by Institute of Mathematics and its Applications
A Bayesian approach to equity valuation
Quaestor Investment Management, River Plate House London EC2M 7TT
Quantitative approaches to equity valuation in common use include methods based upon the dividend discount model, which equates the value of an enterprise to the present value of the dividends that it pays to its owners, and methods which begin from the relation between the current share price and the expected earnings for the next 12 months. A short coming of each of these approaches is that they are forced to take near-term earnings and dividend forecasts as their basic data, when in fact the values they are trying to estimate are dominated by longer-term considerations. As a result, model estimates are often too sensitive to changes in near-term earnings which have little effect on a company's long-term prospects.
In relatively efficient markets, however, the price of a company's shares at any time reflects information which is not readily quantifiable and which may represent a longer-term view. That information can be used to create a stable quantitative valuation model in which the information content of near-term changes is balanced against longer-term considerations via a Bayesian updating procedure.
The structure of the Bayesian valuation model is essentially that of a truncated dividend discount model, where the fair value today is expressed as the present value of a terminal dividend (i.e. the fair value at a future date, discounted to the present), plus the present value of the interim dividends. In the Bayesian model, the terminal dividend is a variable whose probability distribution depends upon the sequence of earnings between the valua tion date and the terminal date.