ATLASMonte Carlo simulation
Investing glossary · Models and probabilities
A method that runs thousands of random scenarios to show the range and likelihood of possible outcomes.
Definition
A Monte Carlo simulation answers "what could happen?" by running a model thousands of times with random inputs. Each run is one possible future; together they form a distribution of outcomes, from which you read ranges and probabilities, such as the chance of ending above today's price.
In investing it is used to project prices and portfolios when no formula gives the answer directly. Its strength is that it shows uncertainty instead of hiding it behind one forecast. Its weakness is that the output is only as good as the assumptions: if the volatility or the shape of returns is wrong, thousands of runs will be consistently wrong.
Formula
S_T = S₀ · exp( (μ − σ²/2)·T + σ·√T·Z ), Z ~ N(0, 1) (repeated thousands of times)
S₀ = today's price, μ = expected annual return, σ = annual volatility, T = time in years, Z = a random draw from a standard normal distribution. Each run draws a new Z.
Worked example
In the ATLAS simulation for NVIDIA (NVDA), priced at $219.06, the median outcome after 21 trading sessions is $219.00. Nine out of ten outcomes land between $182.38 and $262.97, and 50% of the outcomes end above today's price.
The width of the range is the message: the simulation shows how uncertain the next month is, not what will happen.
How ATLAS uses it
ATLAS runs simulations to produce the expected range and the probability of rising on every stock page, and lets you test your own assumptions in the Scenario simulation tool.
Related in ATLAS: Scenario simulation →Browse all ATLAS tools →See it on a real stock
The example above uses these companies. Open their pages for the full numbers:
Frequently asked questions
Does a Monte Carlo simulation predict the future?
No. It shows the spread of outcomes that follow from its assumptions. It tells you what is plausible, not what will happen.
How many simulations are needed?
Enough for the results to stop changing from run to run; thousands are typical. More runs give smoother estimates, not better assumptions.
Related terms
Put the idea to work on your own stocks
Open any company in ATLAS to see its fair value, its odds of rising and its risk numbers side by side.
Try ATLAS →Not investment advice. ATLAS estimates are produced by statistical models from public data; they can be wrong and past performance does not predict future results. Consider your own situation before investing. Disclaimer