Skip to content

Harris And Stevens Forecasting

  • Builds statistical models from past observations (e.g., past sales or passenger counts) to predict future demand.
  • Explicitly accounts for seasonality but assumes stationary demand over time.
  • Simple to implement and used to inform inventory, production, and marketing decisions.

Harris and Stevens forecasting is a method used to predict the future demand for a product or service by applying statistical techniques to historical data to determine likely future demand.

The method gathers historical measurements relevant to demand (for example, counts by time period and contextual attributes) and fits a statistical model to those observations. It explicitly incorporates seasonality effects, making it suitable for cyclical demand patterns. The approach assumes demand is stationary—i.e., that the underlying demand process remains constant over time—which is a core limitation when conditions change. Because it is based on historical data and standard statistical techniques, it is relatively straightforward to implement and can be applied by businesses of varying sizes.

Airlines gather data on past flights (including the number of passengers, the time of year, and the destination) and use that data to create a statistical model that predicts demand for future flights.

Retail stores collect past-sales data (including the number of units sold, the time of year, and the type of product) and use it to build a statistical model that predicts demand for future products.

  • Forecasting inventory levels
  • Scheduling production
  • Planning marketing strategies
  • Applicable across businesses of different sizes and in industries with cyclical demand
  • Relies on stationary demand; the assumption that demand remains constant over time may be invalid in rapidly changing industries.
  • Does not account for changes in consumer preferences or broader market conditions, or for external factors such as economic shifts.
  • Historical data
  • Statistical techniques / statistical model
  • Seasonality
  • Stationary demand
  • Consumer preferences
  • Market conditions