A z score states how far an observation lies from the average of a distribution, scaled by how spread out that distribution is. In markets it can flag unusual returns or accounting ratios. A related but separate tool, the Altman Z score, combines several financial ratios to assess bankruptcy risk for some types of firms.
Key takeaways
- A z of zero means the value equals the mean of the reference sample.
- Large absolute z scores mark outliers, which may be errors, rare events, or regime changes.
- The measure assumes a meaningful mean and standard deviation. Bad samples produce bad z scores.
- Do not confuse a simple statistical z with any branded credit score without reading the model definition.