What are two common methods of demand forecasting?

• A. Time Series Only
• B. Time Series and Causal Models
• C. Qualitative Guesses Only
• D. Random Sampling

Answer: (B)

Two common methods for forecasting demand blend historical patterns with the influence of external drivers. Time series analysis looks at past sales data to uncover patterns that repeat over time—like trends, seasonality, and cycles—and uses that information to project future demand. Techniques range from simple moving averages to more advanced methods like ARIMA or exponential smoothing, depending on the data.

Causal models, on the other hand, try to explain and quantify how external factors affect demand. By relating demand to variables such as price, advertising spend, promotions, macroeconomic indicators, or weather, these models (often regression-based) allow you to estimate how changes in those drivers would change future demand and to run what-if scenarios.

This combination is common because most organizations have both historical sales data and some measurable drivers that influence demand. Using time series captures inherent patterns in the data, while causal models account for external factors that could alter those patterns. In practice, teams often use both approaches together or compare them to improve forecasting accuracy.

Why the other ideas don’t fit as the primary approach: relying solely on qualitative guesses ignores the strong signal in actual data; random sampling isn’t a forecasting method and doesn’t produce future demand projections; and stating only time series omits the valuable information external drivers can provide.