This article is intended for finance professionals curious about integrating advanced statistical techniques into their various processes. The focus is not on the methods and merits of any specific modeling technique but instead focuses more on preparing finance processes for their inevitable use.
The famous quarter of Monaco, a city-state surrounded by France and sitting on the Mediterranean, first lent its name to a secret experiment at Los Alamos Scientific Laboratory back in the 1940s. I am fairly certain that at this point, practitioners in corporate finance were more interested in Le Grand Casino than in the computational experiment with random variables that adopted its name. Something tells me they still may be. As a result this commentary is less about stochastic simulation used by leading finance organizations in forecasting and managing risk and more about the evolution of financial processes required to effectively utilize advanced techniques like Monte Carlo Simulation (MCS).
Demand planning, costing, pricing, strategic planning and valuations are examples of complex models where MCS may be useful. My experience has shown that finance organizations looking to leverage more sophisticated methods and technologies could benefit from a practical roadmap to evolving their processes to the point where these techniques are useful.
The typical evolutionary path for any given financial process follows these four progressive steps:
- Driver Based Models
- Scenario Testing
- Sensitivity Testing
- Simulation Models
Application modernization is a growing area of focus for enterprises. If you’re considering this path to cloud adoption, this guide explores considerations for the best approach – cloud native or legacy migration – and more.
Driver Based Models: These types of models are very common these days and have proved their value in bringing rigor to modeling the complexities of business. Many organizations are still refining their set of drivers to more closely align them with operational reality. Even the exercise of identifying and testing business drivers has proven valuable for business.
Scenario Testing: With efficiency gains driven by technology advancements, financial models are now enabling multiple scenarios and scenario management. For many, producing only one version of the plan is a complicated, lengthy and error prone process preventing them from producing multiple scenarios. Most organizations find their processes and models in this phase of transition. They are working to enable multiple scenarios and changing baskets of drivers to model various potential outcomes.
Sensitivity Testing: By the very nature of having multiple scenarios most organizations probably have some indication of sensitivity. We frequently find scenarios with names like Pessimistic, Most Likely and Optimistic. Assuming your inputs are accurately depicting these scenarios these can provide a rough indication of sensitivity. Enabling more specific sensitivity testing of the various drivers can highlight where the model is most sensitive to changes in input values.
Simulation Models: With each step building on the last a finance organization may now be ready for simulation models. Models, like MCS, consider all possible combinations of driver values and generate a probability distribution describing the possible outcomes. The simulation model enables you to calculate a most likely scenario as well as a set of reasonably probable ones.
Traversing these steps is neither an impossible nor a lengthy task. For finance teams looking to evolve their capabilities, it is all within their grasp. Technology vendors have been enabling these elements for years and with the right guidance and focus your team can find Monte Carlo too.
About the Author:
Erik Duffield is a Managing Director with Perficient a Performance Management Consulting firm. By combining best practices with deep experience and the ability to execute, Perficient enables finance organizations to get “unstuck” from the transaction functions of the role and transition to driving material results in the organization.