Investing with a Stock Valuation Model - Yale University：投资股票的估值模型-耶鲁大学.ppt

Motivation: existing stock valuation models Variants of the Gordon model: too many unrealistic assumptions (e.g., a constant and flat term structure, constant dividend growth forever) Multi-stage dividend/earnings/cashflow discount models: No structural parameterization of the firm’s business No attention paid to how the stock has historically been valued by market Fair values determined by these models are too often below market price. The Bakshi-Chen-Dong (BCD) Model Fundamental Variables: current EPS, expected future EPS, and 30-yr bond yield Firm-specific parameters: EPS growth volatility Long-run EPS growth rate Duration of business-growth cycle Systematic or beta risk of the firm Correlation between the firms EPS and the interest-rate environment 30-yr Treasury yield’s parameters: Its long-run level Interest-rate volatility Duration of interest-rate cycle Comparison The BCD Model Detailed parameterization of EPS processes and interest-rate processParameters to be estimated from past data Closed-form stock valuation formula Past data are used to estimate parametersSo, valuation reflects both past valuation standard for the stock and the stochastic discounting of future prospects The Residual-Earnings Model (e.g., Lee, Meyer and Swaminathan (1998)) Two parameters: beta and dividend-payout ratio No closed-form valuation formula. Requires ad hoc approximation of the stock’s future price at end of forecasting horizon Valuation is independent of past valuation standard for the stock Data I/B/E/S, CRSP, and Compustat Future EPS forecasts: consensus analyst estimates Period covered: Jan. 1979 - Dec. 1996 Stock universe: about 2500 U.S. stocks (mostly large cap) What Constitutes a Good Stock-Selection Measure? Mean-reverting, so that if too low, you can buy the stock, counting on the measure to go back to its norm. Not too persistent, e.g., if book/market ratio is too persistent, you will not want to buy a stock ju