What do alpha and 1 minus alpha represent in this forecasting method?

Conditions

• $\alpha$ is fixed at 0.4 in the worked example.
• $(1-\alpha)$ is calculated as 0.6.
• The weights must sum to 1 for the formula to represent a valid weighted average.

Reasoning, step by step

1. Identify $\alpha$ as the parameter controlling the responsiveness to recent actual data.
2. Identify $(1-\alpha)$ as the parameter controlling the reliance on the previous forecast.
3. Calculate the complementary weight: $1 - 0.4 = 0.6$.
4. Apply these weights to $A_t$ and $F_t$ respectively in the formula $F_{t+1} = \alpha A_t + (1-\alpha)F_t$.

Example

The instructor states that the problem gives alpha 0.4 and computes 1 minus alpha as 0.6. Handwritten annotations show $\alpha = 0.4$ and $(1-\alpha)=1-0.4=0.6$, which are then used to weight the actual demand and forecast in each step.

Common misconceptions

• Assuming $\alpha$ represents the weight given to the previous forecast instead of the actual demand.
• Believing that $\alpha$ and $(1-\alpha)$ can sum to a value other than 1.

Watch the explanation

Explore next

• How do you calculate exponential smoothing forecasts step-by-step?
• What does the smoothing constant alpha represent?