Video summary

22 Quantitative sales forecasting 4 quarter moving average REC

Main summary

Key takeaways

Educational

Summary

The video explains how to use four-quarter moving averages to forecast sales from quarterly data. It covers calculating and centering the averages, adjusting forecasts for seasonal or cyclical patterns, and recognizing the method’s limitations.

Calculating a centered four-quarter moving average

A four-quarter period has no single middle quarter: its midpoint falls between two quarters. To align the average with a specific quarter, the video uses centering:

  1. Add the sales figures for four consecutive quarters to get a four-quarter total.
  2. Move forward one quarter and calculate the next four-quarter total.
  3. Add the two consecutive totals. The result covers eight quarter-values and is centered on the quarter between the two totals.
  4. Divide the sum by eight to calculate the centered four-quarter moving average for that quarter.
  5. Repeat the process to create a moving-average series. Plot it alongside the actual sales data to show the underlying trend.

Forecasting and adjusting for seasonal variation

Draw a line of best fit through the moving averages and extend it into the future to estimate the underlying sales trend. A trend-only forecast may not reflect the fact that some quarters regularly have higher or lower sales.

For each quarter with an available moving average, calculate its variation:

Variation = Actual sales − Moving average

Group the variations by quarter and calculate the average variation for each quarter. This gives the typical seasonal or cyclical adjustment for that quarter.

Adjust the trend forecast using the relevant quarter’s average variation:

  • Add a positive variation when that quarter typically exceeds the trend.
  • Apply a negative variation when it typically falls below the trend.

In the example, the estimated average variation for quarter three is +£51.67 thousand. Adding this to a trend estimate of £216,000 gives a quarter-three forecast of £267,670.

The example’s quarter-one average variation is −£33.4 thousand, so a quarter-one forecast should be reduced by that amount. The same process can be used for quarters two and four.

Limitations

  • Past sales patterns may not be a reliable guide to future sales.
  • Changes in the external environment can make historical trends and seasonal patterns less useful, especially if those changes are difficult to predict.
  • The quality and reliability of the data matter.
  • Forecasts become more uncertain the further into the future they extend.

Speakers and sources

  • Speaker: One presenter or narrator explaining the method.
  • Sources or materials mentioned: An illustrative quarterly sales dataset and table shown in the lesson; no named external sources are identified.

Rate this summary

Your feedback will help improve summaries.

Improve this summary

Reprocess with a stronger model when the summary feels incomplete or inaccurate.

Pro

Translate summary in another language

Pro

Ask questions to this video

Chat for follow-up questions, clarifications, and source-backed answers.

Coming soon

Share this summary

Original video