Estimating Project Volatility for Real Option Valuation

Summary

The discussion considers how to estimate project volatility for real option valuation, in response to concern that several named approaches may produce upward-biased estimates. It presents two broad methods: use outcomes from comparable projects when a sufficiently large and relevant historical sample exists, or use simulation when suitable comparables are unavailable. The answer characterizes comparables as generally preferable in principle, while noting that true project matches are difficult to find in quantity.

The response points to a paper describing a volatility estimation technique, but the document does not explain that technique or provide results validating it. It also does not compare the named approaches, quantify their alleged bias, or give practical simulation steps. The main takeaway is therefore a choice of data source rather than a complete estimation procedure: assess whether project comparables are credible and plentiful, and expect simulation to be used when they are not. Conclusions depend on the quality and availability of project-specific evidence.

Key ideas

• Project volatility for real options can be estimated from comparable project outcomes or through simulation.
• Comparable projects are presented as preferable when the historical sample is sufficiently large and relevant.
• Finding enough genuinely similar projects is described as difficult, so simulation is often used.
• The answer references a proposed technique but does not detail or validate it.

Tags

# What is the best method to compute project volatility in Real Option Valuation?


# What is the best method to compute project volatility in Real Option Valuation?












There are few methods like Copeland-Antikarov, Herath-Park, Cobb-Charnes etc. to compute project volatility, however these methods compute upward biased volatility. What is the best method I could use to compute project volatility for real option valuation?

## Answer by glyphard (score 3, accepted)

https://quant.stackexchange.com/a/1131

There are two main approaches:

- Comparables (depends on having an existing sample of similar project outcomes, which can be difficult to obtai [similar to historical volatility for stocks])

- Simulation (when comparables, or rather enough of them, aren't available)

Here's a link to a paper that provides a technique for estimating volatility for real options:

http://new.vmi.edu/media/ecbu/cobb/EE4902.pdf

Generally comparables is the preferable approach in real options, provided that you have a large enough historical sample. In most cases, however simulation is used as 'true' comparables are notoriously difficult to find en masse.

Shown in full with attribution under the source's licence. Licence: CC BY-SA 4.0 (Stack Exchange)

This summary was written by Stratmill's research agent from the original; it is not a copy of the source.