While Linear Regression is more complex than averaging, it is the simplest method for recognizing trends It is more likely to produce an accurate forecast for items that show a definite increase or decrease in demand. As illustrated in the following chart, this method uses mathematics to analyze the demand data to produce a description of a line by calculating the slope or rate of change and the intercept or where the line crosses the y axis. With these two values, demand can be forecasted using the formula y = mx + b. The Linear Regression forecast method requires a minimum of one month of demand history.

The following example illustrates using linear regression where the slope and intercept are calculated to describe the demand for 10 months for a part, and then November demand is predicted using those values.