Skip to content
Event Studies and Causal Measurement in Finance / Abnormal-return measurement

Abnormal Returns

The earnings date is already outside the estimation window, and the market model is the benchmark. The abnormal return is that model's disturbance evaluated on the announcement days, not on the days that produced α^i\hat\alpha_i and β^i\hat\beta_i.
This note writes the ordinary-least-squares fit, the out-of-sample residual, and the conditional variance that still contains parameter uncertainty. Summing those residuals is the cumulative abnormal return.

Cross-sectional averaging waits until the covariance across firms is stated honestly.

Learning objectives. By the end of this note you should be able to:
  • compute α^i\hat\alpha_i, β^i\hat\beta_i, and σ^εi2\hat\sigma^2_{\varepsilon_i} on an estimation window of length L1L_1;
  • write the event-window abnormal return as the out-of-sample market-model residual;
  • separate the disturbance variance from the sampling variance of α^i\hat\alpha_i and β^i\hat\beta_i;
  • state the null distribution of one event-window abnormal return.

In this lesson

  1. Fitting the market model
  2. The event-window residual
  3. Reading the residual
  4. Conditional variance
  5. Audit on the running announcement

Before you start

The full lesson, with worked examples and exercise solutions, is for members.

Read the full lesson

New to SelfMadeQuant.com? See membership options or browse the Event Studies and Causal Measurement in Finance syllabus.