Definition: A sensitivity analysis is a mathematical formula used in financial modeling to calculate if a target variable is influenced by other outside variables called input variables.
What Does Sensitivity Analysis Mean?
What is the definition of sensitivity analysis? In a more general sense, this is a process used by investors to understand what market conditions and environmental variables will change their prospective investments. For example, investors often use this analysis to test what variables change the stock price of a publicly traded company and to what degree.
Sensitivity analyzes are studied by changing one variable at a time and observing the subsequent effects. By segregating variables through the process of elimination, investors can get a clearer picture of what affects the market prices of stocks. It is also used to gather information and prepare different outcomes in case the data wasn’t interpreted correctly by a system.
Let’s look at an example.
Example
Let’s assume Cathy is a stock analyst at a hedge fund in New York. Her job is to research companies and identify potential investments the fund can explore in the future. Currently, she is testing to see what influences the stock prices of the companies in her portfolio.
She wants to test the level of influence each of the following variables has on the stock price overall.
- Number of Competitors
- Number of Shares Outstanding
- Debt to equity ratio
She looks back through prior years’ financial statements and other historical data to map out a trend with each of these variables. She noticed that as the number of competitors increased over the years, the stock price declined. Likewise, as the DER increased, the stock price also declined. The number of shares didn’t have any effect because no new shares have been issued.
Based on this information, Cathy can make a model to predict how the stock price will be affected by these changing variables in the future.
Summary Definition
Define Sensitivity Analysis: Sensitivity analysis means an evaluation of the amount of error an output holds when it is generated from other data that may also have errors or inaccurate data. In other words, it determines how uncertain an individual can be that the given output is correct considering the sources used.
