One of the most important and challenging aspects of forecasting is handling the uncertainty inherent in examining the future, especially in our modern turbulent times with COVID and the staggering collapse in inflation expectations. Comparing actual outcomes against projections underscores the need to explicitly recognize uncertainty.
For that purpose, Monte Carlo simulations are an extremely effective tool for handling risks and probabilities. This probability simulation is a technique used to understand the impact of uncertainty in financial management, cost, and other forecasting models. It’s mainly used to simulate the various sources of uncertainty that affect the value of business, and to calculate a representative value given these possible values of the underlying inputs. In the simulation, the uncertain inputs are described using probability distributions & your computer therefore randomly draws a number from each input distribution and calculates and saves the result. This is repeated hundreds, thousands, or tens of thousands of times, each called an iteration. When taken together, these iterations approximate the probability distribution of the final result.
How is using probability distribution (aka stochastic models) different to Deterministic models that, for example, use averages as inputs in forecasting? The most notable difference is that deterministic models do not account for real-life stochasticity (i.e., extreme values), which may result in wrong predictions, and therefore wrong decisions. In deterministic models however, the output of the model is fully determined by the parameter values and the initial conditions. Additionally, stochastic models possess some inherent randomness. Our natural world is actually buffeted by stochasticity.
There are very few sources that offer stochastic modeling tools, due to their relative complexity. Among those is Oracle’s Crystal Ball; a spreadsheet-based application for predictive modeling, forecasting, simulation, and optimization, that builds on Monte Carlo modeling tools.
Over the past few months, I have managed to program a similar application using R-language which puts the concepts of Monte Carlo simulation into action. The purpose of my application is to forecast the net profit of a business or product line, based on 4 variable inputs: Sales volume, selling price, unit cost, and fixed cost.
I am giving access to this app for free. To use it, please follow the link below:
https://g0jl5s-samreturns9.shinyapps.io/MonteCarloSimulation/
Enjoy!