Learning Objectives

By the end of this lesson, learners should be able to:

  • Explain regression analysis.
  • Interpret regression equations.
  • Use regression for business prediction.
  • Evaluate model fit conceptually.

What Is Regression Analysis?

Regression analysis estimates the relationship between a dependent variable and one or more independent variables.

Example

Predict sales from advertising expenditure.

Sales = a + b(Advertising)

  • a = intercept,
  • b = expected change in sales for a one-unit increase in advertising.

Worked Example

Suppose the estimated model is:

Sales = 50 + 4(Advertising)

If advertising expenditure is USD 20,000:

Predicted sales = 50 + 4×20 = 130

Interpretation: Expected sales are 130 units (or another defined sales measure).

Business Applications

  • Sales forecasting,
  • Demand prediction,
  • Customer lifetime value estimation,
  • Credit scoring,
  • Pricing analysis.

Interpreting The Coefficient

If b = 4, an additional USD 1,000 in advertising is associated with an average increase of 4 sales units, assuming other factors remain constant.

Goodness Of Fit

R² measures the proportion of variation explained by the model.

  • R² = 0.80 means 80% of sales variation is explained by advertising.

Higher values generally indicate stronger explanatory power, though context matters.

International Example

A global hotel chain predicts monthly bookings using advertising spend, room price, and online review scores across hotels in Paris, Dubai, Singapore, Toronto, and Sydney.

Limitations

  • Correlation does not prove causation,
  • Relationships may change over time,
  • Important variables may be omitted,
  • Extreme values can distort results.

Practical Exercise

Create a simple regression model relating marketing expenditure and sales for five regions and interpret the coefficient.

Learning Materials / Reference Materials

  • Gujarati & Porter. Basic Econometrics.
  • James et al. An Introduction to Statistical Learning.

Lesson Summary

Regression analysis is a powerful predictive tool that estimates relationships between variables and supports business forecasting and decision-making.

PreviousNext