REINZA Logo

Chapter 6: Calculating Forecasts

Predicting the future using time series analysis and linear regression.

About this Lecture

The future is uncertain, but math can help us make educated predictions. In this chapter, we master Time Series Analysis to identify underlying trends and seasonal variations (like holiday sales spikes). We also explore Linear Regression, a powerful tool to predict costs or sales based on historical data patterns.

Study Guide Highlights

1. Components of a Time Series

Any set of data over time can be broken down into four components:

  • Trend (T): The long-term general movement (up or down).
  • Seasonal Variation (S): Regular, predictable short-term patterns (e.g., quarterly).
  • Cyclical Variation (C): Medium-term economic cycles (recession/boom).
  • Random Variation (R): Unpredictable, one-off events.

2. Linear Regression (y = a + bx)

This formula creates a line of best fit through all your data points:

  • y: The dependent variable (what you are trying to predict, e.g., Total Cost).
  • x: The independent variable (the cause, e.g., Units Produced).
  • a: The intercept (Fixed Cost).
  • b: The gradient (Variable Cost per unit).

3. Exam Tips

💡 Centering Moving Averages

If you are calculating a moving average for an EVEN period (e.g., 4 quarters), the result lands "between" quarters. You MUST average two results to "center" it back onto a specific quarter.

💡 Additive vs Multiplicative

Check the question carefully! Additive model means "Trend + Seasonal" (expressed in £/units). Multiplicative means "Trend × Seasonal" (expressed as a %).

Chapter Resources

Download the official study text and working papers for this chapter.

Download Study Text

PDF Document

Right-click and "Save Link As" if download doesn't start.