Monte Carlo simulation

On this page

Learning Outcome Statement

Describe Monte Carlo simulation and explain how it can be used in investment applications.

On the exam: The question gives an asset price, model parameters, a horizon convention and a standardized random draw or simulated output. It asks for a simulated return, a terminal price or an interpretation of the outcome distribution, usually in an MCQ or short table.

You are not trying to predict a certain path. You run the same specified model repeatedly, each time with a random draw from a standardized normal distribution. The draws create different returns and prices, so the output becomes a distribution for investment analysis. Its usefulness depends on the model, parameters, horizon and draws.

Model itemUnit or description
Expected returnDecimal per matching period
VolatilityDecimal per matching period
Standardized normal drawStandard deviations
Initial priceUSD per share
HorizonMatching period

Expected return and volatility must describe the same period as the horizon. The draw supplies direction and size in standard-deviation units, so volatility converts it into return units before the model builds a price.

Keep the two stages distinct as you read the model. First, use the expected return and the scaled draw to find the trial return. Then apply that return to the starting price. This separation makes it easier to check units and catch a sign error before it affects the terminal value.

Part of trialReturn calculationPrice calculation
BaseExpected returnInitial price
Trial inputVolatility-scaled drawOne plus simulated return
OutputSimulated returnTerminal price

Call the simulated simple return $r$. The expected return for the matching period is $\mu$, and the volatility for that period is $\sigma$. The standardized normal draw is $z$. If the asset starts at $P_0$, its terminal price after the period is $P_1$.

The first equation creates one simulated return. The second converts that return into the asset’s terminal price for the one-period simple-return model.

$$ r = \mu + \sigma \times z $$

$$ P_1 = P_0 \times (1+r) $$

The model starts at $\mu$, which provides the center of the return. The term $\sigma \times z$ is the trial-specific shock. A positive $z$ raises $r$, while a negative $z$ lowers it. The price then moves with $r$ because $P_0$ is multiplied by $1+r$.

Read the process as a loop. A draw becomes one outcome, and the same model is run again with another draw.

In each pass, the assumptions stay fixed while the standardized draw changes. That lets you attribute differences across outcomes to the random shock rather than to an unannounced change in the model. Matching the period remains essential: annual volatility cannot be applied directly to a monthly horizon without adjusting the inputs.

Loading diagram…

Monte Carlo is useful when the investment question concerns a range of possible outcomes rather than a point estimate. An analyst can use the resulting distribution to examine possible terminal prices, returns or the chance of crossing a target. The analyst can rerun the specified model with different parameters and compare the resulting distributions.

Use the process picture to keep the order straight: specify the model, generate a draw, calculate its return and price, then repeat. A single result is only one possible path. Comparing many terminal prices shows how the assumptions and shocks combine across trials, which is the basis for interpreting a simulated outcome distribution.

Worked example

You are handed an asset simulation with a one-year horizon. Its initial price, annual expected return, annual volatility and standardized draws are listed below. You need to turn each draw into a simulated return and terminal price, then summarize the resulting price set.

InputValue
Initial price, $P_0$USD 100 per share
Expected return, $\mu$0.08 decimal per year
Volatility, $\sigma$0.2 decimal per year
Horizon1 year
Trial count5 trials
Draws, $z$0, 1, -1, 0.5, -0.5

The bars below show the terminal price produced by each draw. Read them as separate trial outcomes, not as a smooth line between points.

Before calculating, note the direction each input should push the price. The zero draw leaves only the expected return; a positive draw adds a volatility shock, while a negative draw subtracts one. The half-size draws create half the shock of the full-size draws, so the five prices can be checked against this ordering.

Loading diagram…

Each bar corresponds to one listed trial, and the chart shows the same five terminal values in visual form. The horizontal labels identify the trials rather than elapsed time, so do not read the bars as a price path. Use the table next to match each outcome to its exact value and preserve the trial ordering when checking calculations.

TrialTerminal price (USD/share)
1108.00
2128.00
388.00
4118.00
598.00

Treat the listed draws as independent illustrative trials. For each trial, scale the draw by volatility because the draw is in standard-deviation units, add the shock to expected return and apply the resulting simple return to the initial price.

  1. For trial $1$, the shock is $\sigma \times z = 0.2 \times 0 = 0.0000$. The return is $r = 0.08 + 0.0000 = 0.0800$, so $P_1 = 100 \times (1 + 0.0800) = 108.00$ USD per share.

  2. For trial $2$, the positive draw creates a shock of $\sigma \times z = 0.2 \times 1 = 0.2000$. The return is $r = 0.08 + 0.2000 = 0.2800$, so $P_1 = 100 \times (1 + 0.2800) = 128.00$ USD per share.

  3. For trial $3$, the negative draw lowers the return. The shock is $\sigma \times z = 0.2 \times -1 = -0.2000$, so $r = 0.08 + -0.2000 = -0.1200$ and $P_1 = 100 \times (1 + -0.1200) = 88.00$ USD per share.

  4. For trial $4$, the half-size positive draw creates a shock of $\sigma \times z = 0.2 \times 0.5 = 0.1000$. The return is $r = 0.08 + 0.1000 = 0.1800$, so $P_1 = 100 \times (1 + 0.1800) = 118.00$ USD per share.

  5. For trial $5$, the half-size negative draw creates a shock of $\sigma \times z = 0.2 \times -0.5 = -0.1000$. The return is $r = 0.08 + -0.1000 = -0.0200$, so $P_1 = 100 \times (1 + -0.0200) = 98.00$ USD per share.

The trial prices average to USD 108.00 per share, but the spread from USD 88.00 to USD 128.00 tells an investor to examine the outcome distribution rather than rely on one simulated path.

Changed condition

QuantityOriginalLower-return case
Expected return0.080.04
Trial 2 return0.28000.2400
Trial 2 priceUSD 128.00USD 124.00

Before looking at the changed column, predict that every simulated return moves down because only expected return falls, while volatility and every draw stay fixed. Terminal prices move down with those returns, so the distribution shifts lower without changing its volatility-driven spread. Now calculate trial 3 under the changed return before reading on.

Common error

You may copy the standardized normal draw $z$ straight into the simulated return, treating the draw itself as the return. That shortcut feels right because the draw carries the surprise’s direction and size, but it ignores the expected return and volatility that set the model’s return units. When a question supplies expected return, volatility and $z$, scale $z$ by volatility first, then add expected return.

Calculator

Enter the trial terminal prices into one-variable data mode, then retrieve their arithmetic mean.

KeysDisplay
108.00 ENTER108.00
128.00 ENTER128.00
88.00 ENTER88.00
118.00 ENTER118.00
98.00 ENTER98.00
2ND STAT x-bar108.00

Check yourself

Practice

For the base case, what is the terminal price for the trial with z=-1?

Which description best matches Monte Carlo simulation in an investment application?

Put the Monte Carlo steps in order.

  • Specify the model and matching periods
  • Generate a standardized normal draw
  • Calculate the simulated return and price
  • Repeat trials and summarize outcomes

Revision snapshot

Simulated return equals expected return plus volatility times standardized normal draw, $r = \mu + \sigma \times z$.

Terminal price equals initial price times one plus simulated return for the one-period simple-return model, $P_1 = P_0 \times (1+r)$.

Repeat the model with many draws and summarize the distribution of outcomes.

Monte Carlo results depend on the specified model, parameters, horizon and random draws. Do not treat $z$ as the simulated return. On the exam, scale the draw, add expected return, calculate price and interpret the outcome distribution.

Unlock the full lesson and continue reading

Signup to continue reading this lesson and unlimited access questions, flashcards, AI notes, and more

Scan to download app

Scan to download
UNLOCK FREE ACCESS

Practice Questions: Monte Carlo simulation

  1. 1. Northbridge Asset Management is developing a model to forecast next-month equity-index returns. The research team first examines the time-series properties of the index's log price level. It estimates an autoregressive model and conducts augmented Dickey–Fuller tests on both the level and its first difference. The ADF test's null hypothesis is that the tested series has a unit root. The team uses a 5% significance level and assumes that the test specifications include the appropriate deterministic terms and lag structure. The team then compares a linear prediction model with a gradient-boosted tree model. Prediction error is measured using root mean squared error, so lower values indicate better performance. Exhibit 1 reports training and validation results. The model-selection objective is out-of-sample forecasting rather than explanation of historical observations. The team recognizes that a flexible model can fit noise in the training sample and therefore produce substantially worse results on unseen observations. Exhibit 2 summarizes the project's data-handling and validation procedures. Because observations are monthly and forecasts must be generated using only information available at each forecast date, the time ordering of the data is economically relevant. The team considers replacing the random split with a procedure that repeatedly trains on an earlier window and validates on a later window. The macroeconomic database creates an additional concern. Government agencies revise previously published values, but an investor making a historical forecast would have observed only the data vintage released at that time. The current project instead uses the most recently revised value for every historical month. The analyst also labels one sample as the final test sample. After observing its results, however, she repeatedly changes tree depth, learning rate, and the number of boosting iterations. The research director notes that this practice makes the designated test sample part of the model-development process. A new, untouched test period would therefore be required before making a final production decision.

    • 2. Armitage Research is developing a model to forecast an equity-market valuation index and identify firms likely to default within the next 12 months. The valuation index is highly persistent, so the research team first tests whether its level is covariance stationary. The Dickey–Fuller specification has a null hypothesis that the index contains a unit root. Rejection occurs when the test statistic is more negative than the applicable critical value. The relevant output is reported in Exhibit 1. If the unit-root null cannot be rejected, the team will consider transforming the series before using it in a predictive regression. Potential transformations include taking first differences or using continuously compounded changes, depending on the economic interpretation required. The team recognizes that using a nonstationary dependent variable and nonstationary predictors can produce apparently significant relationships that are spurious unless the variables are appropriately modeled as cointegrated. The database contains monthly observations and will be used for genuine out-of-sample forecasting. One analyst proposes randomly allocating observations between training and test samples. Another recommends an expanding-window procedure: estimate the model using only observations available through a particular date, forecast the next period, add that period to the training sample, and repeat. Armitage requires model validation to reproduce the information sequence that would have existed in live implementation. The default classifier's confusion-matrix counts are also included in Exhibit 1. Because defaults are rare, the team evaluates precision separately from overall accuracy. A predicted positive denotes a firm classified as likely to default. The team also checks every predictor's publication date and identifies one variable based on financial statements restated after the date on which the model would have made its prediction. Separately, Armitage uses Monte Carlo simulation to value a European option under a risk-neutral lognormal price process. To improve computational efficiency, an analyst proposes antithetic variates. For every vector of independent standard normal draws, the simulation will also use the vector with all signs reversed. The same payoff discounting and risk-neutral process parameters will be applied to both paths.

      • 3. Priya Shah leads Altair Capital's project to predict rare monthly liquidity-stress events and simulate the future value of a traded asset. The prediction data contain market prices, trading volume, order-imbalance measures, and macroeconomic releases. Each observation is time-stamped according to when the information became available to an investment manager. Shah requires the research team to replicate a feasible historical decision process: a model evaluated at any date may use only information that would have been available by that date. The model-development procedure is summarized in Exhibit 1. The team separates the observations into consecutive training, validation, and final test periods. However, a junior analyst standardizes each feature using its mean and standard deviation calculated over the entire 2016–2023 sample before dividing the data into those periods. The analyst argues that standardization is harmless because no target labels are directly used. Shah notes that even unsupervised preprocessing can leak information when its parameters are estimated using future observations. The team also proposes randomly shuffled cross-validation. Shah is concerned that shuffled folds could train the model on observations occurring after observations in a validation fold. She considers either expanding-window or rolling-window validation, with each validation period occurring strictly after its associated training period. The final 2023 test set is to be used once, after model selection and hyperparameter tuning are complete. Liquidity-stress events are uncommon. Exhibit 1 reports the test-set event frequency and the initial classifier's predictions. The junior analyst emphasizes the model's 95% accuracy. Shah instead examines class-specific measures, including positive-event recall, because failing to identify an actual stress event is economically costly. Recall is defined as true positive predictions divided by the total number of actual positive events. Separately, the team simulates an asset price using the geometric Brownian motion inputs in Exhibit 2. The continuously compounded expected return and volatility are constant, and no cash distributions occur during the one-year horizon. Under the specified process, the expected terminal price is S0e^(μT), while the median terminal price is S0e^[(μ − 0.5σ²)T]. Random simulation error is assumed negligible for selecting the closest answer because 100,000 independent paths are used.

        • 4. Northlake Research develops models for a lender's investment-risk group. The first project forecasts the monthly change in a credit-market stress index. The analyst initially estimates an autoregressive model using the index level, but the estimated first-order coefficient is close to one. She performs augmented Dickey–Fuller tests on both the level and first-differenced series and reviews a Ljung–Box test of the residuals from the differenced model. The results are shown in Exhibit 1. All tests use a 5% significance level. The analyst then designs the model-validation process. Because observations are monthly and the model will predict future conditions using only information available at the forecast date, the validation process must preserve chronological order. The research director rejects any procedure that allows information from later months to enter the estimation sample used to forecast an earlier month. A final untouched test period will be retained after model selection. A second project classifies credit applications as positive or negative using a labelled data set. Exhibit 2 reports the confusion matrix from a genuinely out-of-sample period. A positive prediction means the model recommends approval. The research team evaluates precision because unnecessary positive predictions create costly manual investigations. It evaluates recall separately because missed positive cases have a different economic cost. The team compares two supervised classification models for production. A gradient-boosted tree achieves an out-of-sample area under the receiver operating characteristic curve of 0.82. A regularized logistic regression achieves an out-of-sample area of 0.76. Both results are stable across chronology-preserving validation windows and both exceed the lender's minimum required AUC of 0.74. The lender's model-governance policy requires the deployed model to provide a reproducible, directionally interpretable reason for each approval decision. Compliance staff must be able to trace the effect of each retained input directly from documented model parameters without relying on a separate approximation tool. The outcome variable is labelled, and management requires a direct estimate of approval class rather than a segmentation of similar applicants. The analyst must therefore consider predictive performance, supervision type, and interpretability together rather than selecting a model solely because it has the highest AUC.

          • 5. Ardent Systematic Strategies is developing a model to forecast whether an equity index will rise or fall during the next month. The research team combines valuation spreads, macroeconomic observations, and issuer-level fundamental ratios. The investment process requires that every predictor used for a historical forecast must have been observable at the date on which that forecast would have been made. A researcher first models an aggregate valuation spread as the first-order autoregressive process reported in Exhibit 1. The innovation is white noise with a constant finite variance. Before using the spread in the forecasting system, the researcher must assess the long-run behavior implied by the estimated intercept and autoregressive coefficient. The team then compares two machine-learning models. Model M is a penalized linear regression estimated using the LASSO objective. Model N is an ensemble of sequentially fitted classification trees. Because the investment objective is directional, the target variable equals one when the next-month index return is positive and zero otherwise. The team will evaluate classification performance using measures appropriate for a binary outcome rather than treating the target as a continuous return. The current model-validation procedure randomly allocates observations from the complete history across training and validation samples. A senior researcher objects that this procedure can allow later market regimes to influence the selection of a model intended for an earlier forecast date. She proposes an expanding-window design in which each validation period occurs strictly after its corresponding training period. Hyperparameters would be selected using these validation windows, followed by a single evaluation using a previously untouched test sample. The team also reviews the data controls in Exhibit 2. The existing database stores point-in-time fundamental values corresponding to their original public-release dates. A vendor offers revised historical values that correct subsequent reporting errors, but those revisions were not known to market participants on the original dates. The senior researcher warns that replacing the point-in-time records with revised values could improve backtested results without improving a deployable forecasting process.

            • 6. Aster Credit Analytics is developing a model to forecast monthly changes in an investment-grade corporate credit spread. The available database covers January 2012 through December 2022 and includes spread levels, macroeconomic variables, market liquidity measures, and sentiment extracted from issuer filings and earnings-call transcripts. The sentiment observations are derived from unstructured text, whereas spreads and macroeconomic variables are structured numerical data. Before estimating a time-series model, an analyst performs augmented Dickey–Fuller tests. The null hypothesis for each test is that the tested series has a unit root. Exhibit 1 reports the results for the spread level and its first difference. The analyst then estimates a first-order autoregressive model for monthly spread changes. All values in the model are measured in percentage points, and the forecast requested is conditional on the most recently observed spread change. Aster also compares a random forest with ridge regression. The random forest is more flexible and achieves substantially higher fit in the training sample, but the investment team is concerned that this flexibility may capture noise rather than a stable predictive relationship. Exhibit 2 reports training and test R-squared values. The team's objective is to select the model expected to perform better on new observations, not the model that describes the training sample most closely. The original project design assigns observations from 2012–2020 to training, 2021 to validation, and 2022 to final testing. Hyperparameters are selected using the validation period, and the test period is intended to remain untouched until the final model evaluation. One analyst instead standardizes every numerical predictor using the mean and standard deviation calculated from the full 2012–2022 sample before creating the three chronological subsets. The analyst also proposes repeatedly changing hyperparameters after reviewing the 2022 test results. The project leader states that preprocessing parameters should be estimated using training observations only and then applied without re-estimation to validation and test observations. Because the observations form a time series, the leader also prefers rolling or expanding-window validation to randomly shuffled cross-validation. This approach preserves temporal order and prevents later observations from supplying information to models trained as if only earlier data were available.

              • 7. Ardent Data Lab develops analytical tools for a commercial lender and an asset-management affiliate. The lender has asked the team to construct a classification model that identifies borrowers likely to default. The full data set contains completed loan outcomes, including both defaults and non-defaults. The team creates training, validation, and test samples and intends the test sample to represent previously unseen observations. Exhibit 1 describes the preprocessing and model-selection sequence. Ardent's governance policy requires that information from the final test sample not influence parameter estimation, feature engineering, hyperparameter selection, or preprocessing estimates. The lender is particularly concerned about failing to identify borrowers who will actually default. Management accepts that this priority may result in some creditworthy borrowers being classified as risky. Exhibit 2 provides the final confusion matrix. The team considers accuracy, precision, and recall but must select the metric most directly aligned with management's stated cost asymmetry. All observations belong to exactly one actual and one predicted class. The asset-management affiliate separately asks Ardent to simulate a security price under geometric Brownian motion. Exhibit 3 provides the initial price, continuously compounded expected return, volatility, and horizon. Under the assumed process, the terminal price is lognormally distributed. The expected terminal price differs from the median because the distribution is positively skewed. The analyst is specifically asked for the median, which is obtained by setting the standard normal innovation in the log-price equation equal to zero, and separately for the variance, which reflects the dispersion of the lognormal distribution rather than its central tendency. No dividends, taxes, or currency conversions apply. Finally, the affiliate evaluates a trading signal using the results in Exhibit 4. The null hypothesis is that mean monthly abnormal return equals zero, and the alternative is that it differs from zero. Statistical significance is assessed at the stated significance level. Economic significance is assessed after considering the transaction cost required to implement the signal each month. The transaction-cost estimate excludes management fees but includes expected bid–ask spreads, commissions, and market impact. The conclusions regarding the credit model, classification metric, simulated price, price variance, and trading signal must each be determined independently.

                Flashcards: Monte Carlo simulation

                1. 1. What is a Monte Carlo simulation?

                  Show answer

                  The repeated generation of risk factors using assumed probability distributions to generate a distribution of security values.

                2. 2. What is a major limitation of Monte Carlo simulation?

                  Show answer

                  The results are only as accurate as the assumptions about the probability distributions used.

                3. 3. What is a major advantage of Monte Carlo simulation over historical simulation?

                  Show answer

                  It can test theoretical scenarios that fall outside the range of actual historical data.

                4. 4. What is a key limitation of bootstrap resampling in simulation?

                  Show answer

                  Its inputs are limited entirely by the distribution of actual historical outcomes.

                5. 5. In valuation, what technique randomly selects values for key variables from probability distributions to generate a range of outcomes?

                  Show answer

                  Simulation.

                Everything you need for CFA prep

                Get full Oncourse access with lessons, practice questions, flashcards and AI study tools.

                GET STARTED FOR FREE
                75% OFFLimited time offer
                GET 75% OFF