Abstract:
In all investment decisions it is important to determine the degree of uncertainty associated with the valuation of a company. We propose an original and robust methodology to company valuation which replaces the traditional point estimate of the conventional Discounted Cash Flow (DCF) with a probability distribution of fair values. It hinges on two main ingredients: an econometric model for the company revenues and a set of firm-specific balance sheet relations that are estimated using historical data. The effectiveness and scope of our methodology are explored through a series of statistical exercises on publicly traded U.S. companies. We show that an uncertainty-adjusted indicator of mispricing, derived from the fair value distribution, is capable of predicting future abnormal returns. Then, we construct a new long-short valuation factor and we test that it is not redundant for describing average returns when used to augment traditional market factor models.