A Hierarchy of Approaches  761
nor required, you can choose from the following three variations of a standard 
DCF approach, depending on the level of uncertainty:
1. Single-path DCF valuation. When little uncertainty exists about future 
outcomes or when uncertainty is evenly spread around the expected 
outcomes, use a standard, single-path DCF analysis based on point esti-
mates of future cash flows.
2. Scenario-based DCF. When significant uncertainty exists, especially when 
there is a possibility of much more upside than downside (or vice versa) in 
future cash flows, it is best to model future outcomes in two or more scenar-
ios that capture the variation in the paths of future cash flow. This approach 
is easy to apply in, for example, valuing corporate or business strategies.
3. Stochastic simulation DCF. If you have reliable estimates about the un-
derlying probability distributions of cash flows into the future, such as 
mean, standard deviation, and possibly skewness, it may be worthwhile 
to use a stochastic simulation DCF approach. In this approach, future 
cash flow paths are explicitly modeled and valued in a stochastic simu-
lation. Because this approach is complex and requires voluminous data, 
applications are mostly restricted to specific industries, such as the valu-
ation of insurance companies, and commodity-based businesses.
When managerial flexibility is called for, you need one of the following 
contingent valuation approaches, selected according to the amount of infor-
mation available:
• Decision tree analysis (DTA). If there is limited information about the dis-
tribution of future cash flow paths and the decisions that management 
can take depending on these cash flows, use a decision tree analysis. 
As the following sections discuss, it builds on scenario DCF valuation 
and is straightforward and transparent. DTA is especially effective for 
valuing flexibility related to technological risks that are not priced in 
the market, such as investments in research and development (R&D) 
projects, product launches, and plant-decommissioning decisions.
• Real-option valuation (ROV). If you have reliable information about the under-
lying probability distributions of future cash flow paths, like those required 
for stochastic simulation, ROV could provide better results and insights. 
However, it requires sophisticated, formal option-pricing models that are 
harder for managers to decipher than DTA. The ROV approach is best suited 
to decisions in commodity-based businesses, such as investments in oil and 
gas fields, refining facilities, chemical plants, and power generators, because 
the underlying commodity risk is priced in the market.2
2 See, for example, E. S. Schwartz and L. Trigeorgis, eds., Real Options and Investment under Uncertainty: 
Classical Readings and Recent Contributions (Cambridge, MA: MIT Press, 2001); T. Copeland and V. An-
tikarov, Real Options: A Practitioner’s Guide (New York: Texere, 2003); or L. Trigeorgis, Real Options: 
Managerial Flexibility and Strategy in Resource Allocation (Cambridge, MA: MIT Press, 1996).