Date of Award
Doctor of Business Administration (DBA)
Economics and Finance
In valuation research, two modeling approaches that have become prominent are those based on the Residual Income Model (RIM) and those based on the G. Feltham-James A. Ohlson framework. Ohlson (1995) develops a valuation model which links a firm's fundamental value to the book value of equity, earnings and other relevant information. Feltham and Ohlson (1995) extend the Ohlson (1995) model to incorporate growth and conservative accounting.
This study provides an evaluation of the Feltham-Ohlson (1995) model assuming market inefficiency. Analyst forecast data are obtained from the international I/B/E/S files. Financial information and share prices are obtained from the Compustat Database. Canadian T-bill rates and exchange rates are obtained from the International Financial Statistics database. All variables are scaled by the market value of equity at fiscal year end to mitigate for heteroscedasticity. Financial firms are excluded. Following Myers (1999), the discount rate is measured as the sum of the Canadian T-bill rate and the firm's industry risk premium. Panel data methodology with lagged values is used to determine the parameters of the linear dynamics equations.
Net operating assets are found to have a negative relationship with abnormal earnings. For the firms in the sample, net operating assets are diminishing over the time period 1990–1998. Managers are selling off assets or they are not making investments sufficient to offset the effects of depreciation.
For every year from 1989 to 1998, four portfolios are formed based on the V/P ratio where V is the predicted value of the firm based on the Feltham-Ohlson (1995) model and P is the market value at fiscal year end. There is a statistically significant difference in the one year returns on low (V/P) portfolios and high (V/P) portfolios. Noise traders acting on pseudo signals continue to invest in overvalued stocks. Professional arbitrageurs are unable to restore equilibrium because of their limited wealth and time horizons.
The differences in the equally weighted 36 month return for the low (V/P) and the high (V/P) portfolios are not statistically significant, indicating that investors become less optimistic about overvalued stocks within 36 months.
Pillay, Colin Anthony, "" (2004). Dissertation. 639.