Monte Carlo Valuation: Using Probability Distributions Instead of Single-Point Estimates
Monte Carlo simulation replaces single-point DCF estimates with probability distributions, producing a range of intrinsic values. Learn how this method improves investment decisions under uncertainty.
Why single-point estimates are misleading
A traditional DCF model produces a single intrinsic value based on a single set of assumptions about revenue growth, margins, and discount rates. But the future is uncertain, and small changes in these inputs can produce wildly different outputs. Monte Carlo simulation addresses this by running thousands of scenarios, each drawing randomly from probability distributions you define for each input. The result is not a single number but a distribution of possible values, giving you a much clearer picture of the risk and reward.
How Monte Carlo valuation works
- Define probability distributions for key inputs: revenue growth rate, operating margin, capital expenditure, and discount rate
- Run thousands of simulated DCF calculations, each drawing a random combination of inputs
- Plot the resulting intrinsic values as a histogram to see the full range of outcomes
- Calculate the probability that the stock is undervalued at the current market price
- Use the median or expected value as your base case and the distribution tails to understand downside risk
Monte Carlo reframes the investment question from "what is this stock worth?" to "what is the probability that this stock is undervalued at the current price?" This probabilistic framing is far more honest about the uncertainty inherent in any valuation.
Start with a solid base case
Before running simulations, build a fundamental DCF model to understand the key value drivers and which inputs matter most.
FAQs
Do I need special software for Monte Carlo valuation?▼
You can run basic Monte Carlo simulations in a spreadsheet using random number functions and data tables, though dedicated tools or Python scripts make it easier to run thousands of iterations and visualize results.
How do I choose the right probability distributions for inputs?▼
Base them on historical data and reasonable expectations. For example, if a company has grown revenue between 5% and 15% over the past decade, a normal distribution centered at 10% with a standard deviation of 3% is a reasonable starting point. Use wider distributions when uncertainty is higher.
Is Monte Carlo valuation more accurate than a standard DCF?▼
It is not necessarily more accurate in predicting the exact intrinsic value, but it is more informative because it quantifies uncertainty. Knowing there is a 70% chance the stock is undervalued is more useful than a single estimate that implies false precision.
Related
Intrinsic Investor is for education and research only. Not financial advice.