What is the exponential smoothing forecast formula shown in the video?

Conditions

• Used recursively over discrete time periods.
• Requires a starting forecast value.
• In the example, alpha is supplied as 0.2.

Reasoning, step by step

1. Identify the general formula displayed on the worksheet.
2. Recognize that $F(t)$ is the current forecast and $F(t-1)$ is the previous forecast.
3. Understand that the term ($A(t-1) - F(t-1)$) represents the forecast error from the previous period.
4. Note that alpha scales this error to determine the adjustment size.
5. Combine these components into the recurrence relation.

Example

The worksheet displays the formula $F(t) = F(t-1) + \alpha \cdot (A(t-1) - F(t-1))$. The speaker explains that the forecast for the current period equals the previous forecast plus alpha times the previous period's forecast error.

Common misconceptions

• Thinking the formula uses the current actual value $A(t)$ instead of the previous actual $A(t-1)$.
• Confusing the smoothing constant alpha with the forecast error itself.

Watch the explanation

Explore next

• How do you initialize the first forecast when there is no previous period?
• What does alpha represent in this forecasting method?
• How is exponential smoothing implemented as an Excel formula?